Comparison of Mold Active Triazoles as Primary Antifungal Prophylaxis in Patients with Newly Diagnosed Acute Myeloid Leukemia in the Era of Molecularly Targeted Therapies

Background. Multiple factors influence the choice of primary antifungal prophylaxis (PAP) in patients with acute myeloid leukemia (AML) undergoing remission induction chemotherapy (RIC) given the recent incorporation of targeted leukemia therapies into these regimens. Methods. We evaluated the incidence and characteristics of breakthrough invasive fungal infections (bIFI) in 277 adult patients with newly diagnosed AML undergoing RIC with high-intensity, or low-intensity venetoclax-containing therapy. Patients receiving posaconazole (PCZ), voriconazole (VCZ), or isavuconazole (ISA) for > 5 days as PAP during RIC were included. Echinocandin use prior to, but not concomitantly with, the PAP azole was allowed. IFI (modified EORTC/MSG criteria) occurring after > 5 days of continuous azole exposure or within 14 days of discontinuation were considered bIFI. Results. Proven or probable bIFI were observed in 11 patients (4%). The incidence of bIFI was 2.9% for PCZ, 4.8% for VCZ, and 5.7% for ISA (P = .55). In total, 161 patients (58%) received echinocandin prophylaxis prior to azole initiation. Neither echinocandin exposure nor chemotherapy intensity impacted bIFI rate. Patients with bIFI had a lower rate of absolute neutrophil count recovery > 1000 cells/µL (64% vs 90%, P = .021) or complete remission (CR; 18% vs 66%, P = .002) after RIC. Thirty-eight patients (14%) discontinued PAP due to toxicity, most often hepatotoxicity. Discontinuation due to hepatotoxicity was similar among azoles (PCZ: 13%; VCZ: 15%; ISA: 13%). Conclusions. The rate of bIFI is low during RIC in patients with newly diagnosed AML receiving any of the mold-active triazoles as PAP. Neutrophil recovery and achievement of CR are important for bIFI risk.