Systematic Anti-Mold Prophylaxis Results in Very Low Incidence of Invasive Fungal Infections in Patients Undergoing Allogeneic Hematopoietic Cell Transplantation

Abstract

Recipients of Cord blood (CBT) T-cell depleted (TCD) and mismatched unrelated (URD) donors allografts are at high risk for invasive mold infections (IMI). The impact of mold-active prophylaxis (PPX) on the incidence and outcomes of mold infections is not well defined. Retrospective review of 988 consecutive adults who underwent alloHCT in our center from 2008 through 2014. Standard PPX consisted of micafungin 150mg IV daily from admission to Day (D) +7 post HCT followed by voriconazole through at least D+100. PPX was continued beyond D+100 in pt with GVHD or use of corticosteroids. Patients (pts) with intolerance or contraindication to voriconazole received posaconazole or micafungin. Pts with IMI were identified through microbiology and pathology records. Cases meeting criteria for proven or probable according to EORTC-MSG criteria are included in the analysis. Median age at HCT was 54 years (range 18-76). The most common diagnoses were AML (n = 351, 36%) and lymphoma (n = 214, 21%). Matched-related and URD were used in 686 (69%) pts, mismatched URD and CBT in 142 (14%) and 154 (16%) pts, respectively. Ex-vivo TCD was used in 484 (49%) pts. 63% pts received myeloablative conditioning. Twenty-four pts (2.4%) developed IMI at a median of 203 days (range 7-1327) after HCT. 19 (79%) cases were probable and 5 (21%) proven. Microbiological diagnosis was established in 10 cases (aspergillus spp=5, mucor=5). Thirteen (54%) pts were diagnosed with probable IFI by positive Beta-D-Glucan assay, compatible radiological findings and host factors. Pneumonia was the most presentation (n = 21, 88%). Fourteen (58%) pts received systemic corticosteroids within 30 days prior to diagnosis, including 12 for GVHD. 14 of the 24 pts (58%) were not on the planned anti-mold prophylaxis at the time of IFI, due to azole-related toxicity (n = 7), drug allergy (n = 1) or patient preference (n = 1). 5 pts were off prophylaxis as they were beyond day +100 and had no additional risk factors. No risk factors for development of IFI were identified, due to low number of events. Subgroup analysis identified HCT populations with very low risk of IFI including AML in CR1 (1/248, 0.4%) and myeloproliferative neoplasms (0/58), and others with higher risk such as CBT (5/154, 3.2%) and mismatched URD (6/142, 4.2%). In the TCD-PBSC setting, 12 of 484 pts (2.5%) developed an IFI. The 12 weeks and 1 year overall survival after IFI was 42% (95%CI 32-52) and 29% (95%CI 19-38), respectively (Figure). All 5 pts with Mucor died at a median of 18 days (1-146) of diagnosis (4/5 related to IMI). 1) In this large and heterogeneous cohort including pts at high risk for mold infection who received mold active prophylaxis, we observed low incidence (2.4%). 2) IMI associated mortality remained high, highlighting the need for improved diagnostic techniques and therapeutic strategies.