Outcomes by MIC Values for Patients Treated with Isavuconazole or Voriconazole for Invasive Aspergillosis in the Phase 3 SECURE and VITAL Trials

This pooled analysis evaluated the relationship of isavuconazole and voriconazole MICs of Aspergillus pathogens at baseline with all-cause mortality and clinical outcomes following treatment with either drug in the SECURE and VITAL trials. Isavuconazole and voriconazole may have had reduced efficacy against pathogens with drug MICs of ≥16 µg/ml, but there was no relationship with clinical outcomes in cases where the MIC was <16 µg/ml for either drug.

All-cause mortality (ACM) through day 42 was assessed in all patients with proven, probable, or possible IFD who received at least one dose of study drug (Table 1; seen also Table S1 in ths supplemental material). Among isavuconazoletreated patients (Aspergillus spp. only), the ACM rates associated with pathogens having drug MICs of Յ1 g/ml were 12.1% (4/33; CLSI) and 9.7% (3/31; EUCAST), whereas the ACM rates associated with pathogens having drug MICs of Ͼ1 g/ml were 12.5% (2/16; CLSI) and 16.6% (3/18; EUCAST). Among voriconazole-treated patients (Aspergillus spp. only), the ACM rates associated with pathogens having drug MICs of Յ1 g/ml were 23.5% (4/17; CLSI) and 27.8% (5/18; EUCAST), whereas the ACM rates associated with pathogens having drug MICs of Ͼ1 g/ml were 60% (3/5; CLSI) and 50% (2/4; EUCAST). With the exception of one patient in the isavuconazole group who was infected with multiple fungal species, all patients in both treatment groups infected with Aspergillus spp. with drug MICs of Ն16 g/ml (n ϭ 2 for isavuconazole by either CLSI or EUCAST; n ϭ 1 for voriconazole by EUCAST only) died by day 42. For patients with Aspergillus-only infections, ACM at day 42 was generally higher with voriconazole than with isavuconazole ( Fig. 1) ( Table 1; see also Table S1); however, the study was underpowered to enable a statistical comparison between the groups.
The overall response (composite of clinical, mycological, and radiological [where relevant; see Table S2] responses at end of treatment [EOT] in patients with proven or probable IFD, as assessed by blind-data review committee) (21) was also investigated. Among isavuconazole-treated patients infected with pathogens (Aspergillus spp. only) with drug MICs of Յ1 g/ml, overall successful responses were observed in 45.5% The denominator represents the number of patients whose isolates had that drug MIC (where patients had multiple isolates, the isolate with the highest baseline drug MIC was used); the numerator denotes the number of patients who died. The outcome for a patient whose last known survival status was determined before day 42 or was missing and whose last assessment day was before day 42 was treated as representing death. c Data include Lichtheimia corymbifera (n ϭ 2 patients), Fonsecaea monophora, Chaetomium brasiliense, and Rhizopus oryzae. d Data include Penicillium piceum.
These analyses of primarily wild-type isolates from the SECURE and VITAL trials suggest that a clearly defined relationship between in vitro susceptibility and clinical outcomes may not be detectable below the epidemiological cutoff value of Յ1 g/ml (CLSI) (22) or Յ2 g/ml (EUCAST) for most common Aspergillus spp. (23). Moreover, this analysis was limited by the small number of isolates, which prevented statistical analysis of the relationship between susceptibility and outcomes and precluded comments on suboptimal outcomes associated with high MICs. Using clinical data to verify the adequacy of current breakpoints would require much larger studies and enrollment of patients with infections by strains with drug MICs higher than the breakpoint, the latter of which would not be ethically tenable.

SUPPLEMENTAL MATERIAL
Supplemental material for this article may be found at https://doi.org/10.1128/AAC .01634-18.

ACKNOWLEDGMENTS
Isavuconazonium sulfate was codeveloped by Astellas Pharma Global Development, Inc., and Basilea Pharmaceutica International Ltd. Editorial assistance was provided by Barrie J. Anthony and Ed Parr of Envision Scientific Solutions, funded by Astellas Pharma, Inc. We are grateful for the contributions of the investigators and staff who conducted the SECURE and VITAL trials and to the patients who volunteered for this study. This analysis was funded by Astellas Pharma, Inc. D.R.A. reports grant support and personal fees for consultancy from Astellas, received during the conduct of the study, and grants from Merck, Cidara, Synexis, Amplyx, Actelion, Theravance, Paratek, Medicines, Geom, Melinta, and Zavante, received outside the submitted work. M.A.G. reports grant support and personal fees for speaking from Astellas, received during the conduct of the study, and grants from Pfizer, grants and personal fees from Meiji Pharmaceuticals, grants from Viamet, and grants and personal fees from Cidara, received outside the submitted work. P.K.M. has nothing to disclose. L.L.K., Q.L., and C.L. are employees of Astellas Pharma Global Development, Inc. M.E.J.