MICs measured using CLSI yeast nitrogen base (YNB) medium instead of CLSI RPMI medium for C. neoformans were evaluated. CLSI RPMI medium ECVs for distributions originating from at least three laboratories, which included >95% of the modeled WT population, were as follows: fluconazole, 8 g/ml (VNI, C. gattii nontyped, VGI, VGIIa, and VGIII), 16 g/ml (C. neoformans nontyped, VNIII, and VGIV), and 32 g/ml (VGII); itraconazole, 0.25 g/ml (VNI), 0.5 g/ml (C. neoformans and C. gattii nontyped and VGI to VGIII), and 1 g/ml (VGIV); posaconazole, 0.25 g/ml (C. neoformans nontyped and VNI) and 0.5 g/ml (C. gattii nontyped and VGI); and voriconazole, 0.12 g/ml (VNIV), 0.25 g/ml (C. neoformans and C. gattii nontyped, VNI, VNIII, VGII, and VGIIa,), and 0.5 g/ml (VGI). The number of laboratories contributing data for other molecular types was too low to ascertain that the differences were due to factors other than assay variation. In the absence of clinical breakpoints, our ECVs may aid in the detection of isolates with acquired resistance mechanisms and should be listed in the revised CLSI M27-A3 and CLSI M27-S3 documents.

t Clinical breakpoints (CBPs) are not available for the Cryptococcus neoformans-Cryptococcus gattii species complex. MIC distributions were constructed for the wild type (WT) to establish epidemiologic cutoff values (ECVs) for C. neoformans and C. gattii versus amphotericin B and flucytosine. A total of 3,590 amphotericin B and 3,045 flucytosine CLSI MICs for C. neoformans (including 1,002 VNI isolates and 8 to 39 VNII, VNIII, and VNIV isolates) and 985 and 853 MICs for C. gattii, respectively (including 42 to 259 VGI, VGII, VGIII, and VGIV isolates), were gathered in 9 to 16 (amphotericin B) and 8 to 13 (flucytosine) laboratories (Europe, United States, Australia, Brazil, Canada, India, and South Africa) and aggregated for the analyses. Additionally, 442 amphotericin B and 313 flucytosine MICs measured by using CLSI-YNB medium instead of CLSI-RPMI medium and 237 Etest amphotericin B MICs for C. neoformans were evaluated. CLSI-RPMI ECVs for distributions originating in >3 laboratories (with the percentages of isolates for which MICs were less than or equal to ECVs given in parentheses) were as follows: for amphotericin B, 0.5 g/ml for C. neoformans VNI (97.2%) and C. gattii VGI and VGIIa (99.2 and 97.5%, respectively) and 1 g/ml for C. neoformans (98.5%) and C. gattii nontyped (100%) and VGII (99.2%) isolates; for flucytosine, 4 g/ml for C. gattii nontyped (96.4%) and VGI (95.7%) isolates, 8 g/ml for VNI (96.6%) isolates, and 16 g/ml for C. neoformans nontyped (98.6%) and C. gattii VGII (97.1%) isolates. Other molecular types had apparent variations in MIC distributions, but the number of laboratories contributing data was too low to allow us to ascertain that the differences were due to factors other than assay variation. ECVs may aid in the detection of isolates with acquired resistance mechanisms.

Cryptococcus neoformans and Cryptococcus gattii are responsible globally for almost one million cryptococcosis cases yearly, mostly in immunocompromised patients, such as those living with HIV. Infections due to C. gattii have mainly been described in tropical and subtropical regions, but its adaptation to temperate regions was crucial in the species evolution and highlighted the importance of this pathogenic yeast in the context of disease. Cryptococcus gattii molecular type VGII has come to the forefront in connection with an on-going emergence in the Pacific North West of North America. Taking into account that previous work pointed towards South America as an origin of this species, the present work aimed to assess the genetic diversity within the Brazilian C. gattii VGII population in order to gain new insights into its origin and global dispersal from the South American continent using the ISHAM consensus MLST typing scheme. Our results corroborate the finding that the Brazilian C. gattii VGII population is highly diverse. The diversity is likely due to recombination generated from sexual reproduction, as evidenced by the presence of both mating types in clinical and environmental samples. The data presented herein strongly supports the emergence of highly virulent strains from ancestors in the Northern regions of Brazil, Amazonia and the Northeast. Numerous genotypes represent a link between Brazil and other parts of the world reinforcing South America as the most likely origin of the C. gattii VGII subtypes and their subsequent global spread, including their dispersal into North America, where they caused a major emergence.

The data available in the literature concerning Cryptococcus gattii in vitro antifungal susceptibility are contradictory. We have analyzed the activity of eight antifungal agents against 23 C. gattii clinical isolates and compared the susceptibility profiles with those of C. neoformans. MIC analysis (mg/L) revealed that C. gattii isolates were more susceptible to amphotericin B and flucytosine than were C. neoformans isolates. Fluconazole and other azole compounds showed high MIC values for C. gattii. Posaconazole displayed good activity. Further studies are required to ascertain the predictive value of the in vitro data presented here.

3.1% of HIV-infected inpatients with CD4 <200 cells/μl without symptomatic meningitis had cryptococcal antigenemia in São Paulo, suggesting that routine CRAG screening may be beneficial in similar settings in South America. Our study reveals another targeted population for CRAG screening: hospitalised HIV-infected patients with CD4 <200 cells/μl, regardless of ART status. Whole blood CRAG LFA screening seems to be a simple strategy to prevention of symptomatic meningitis.