Delaying the Start of Triazole Prophylaxis 7 Days After Allogeneic HCT Does Not Affect Outcomes, Including Risk of Invasive Fungal Infection

Introduction: Antifungal prophylaxis with azoles is considered standard in patients undergoing allo-HCT. However, azoles have significant interactions with immunosuppressive drugs used for graft-versus-host disease (GVHD) prophylaxis. Since the risk of GVHD is increased in the setting of sub-therapeutic immunosuppressant levels, we changed our standard practice from initiating the triazole on day +3 to day +7 after allo-HCT in order to make the change at steady state and limit the occurrence of non-therapeutic levels.

Methods: We retrospectively evaluated 196 patients treated from Jan 2009 to Nov 2011 to assess the impact of delaying the start of azole antifungals on immunosuppressive levels, acute GVHD, and rate of invasive fungal infections (IFI). Micafungin was given from admission until azole initiation. Patients received voriconazole (n=162), posaconazole (n=17), fluconazole (n=9) or were maintained on micafungin (n=8).  Stem cell sources included PBSCT from related/unrelated donors (n=139) and cord blood grafts (n=57). GVHD prophylaxis was with tacrolimus (tacro) + sirolimus (siro) + methotrexate (MTX) (n=76), tacro + MTX (n=42), cyclosporine (CSA) + mycophenolate mofetil (MMF) (n=57), tacro + MMF (n=9), tacro + siro (n=7), CSA + MTX (n=3), CSA (n=1), siro + MMF (n=1). Results were analyzed by intent-to-treat (ITT) for patients treated before (day+3 switch, n=69) or after February 2010 (day+7 switch, n=127) as well as by actual day of switch + 1 (day+2,3,4: n=49 vs. day+6,7,8: n=70). Cumulative incidence (CI) functions estimated day 200 grade II-IV aGVHD. Gray’s test was used to compare aGVHD incidence for individuals in the day +3 and day+7 treatment categories for both the ITT and as-treated analyses.  

Results: Median levels two weeks after HCT were therapeutic in 19/19 (100%), 50/52 (96%), 31/32 (97%) patients on CSA, Tacro, and Siro, respectively, prior to Feb 2010 and in 43/44 (98%), 82/83 (99%), 47/52 (90%), for CSA, Tacro, and Siro, respectively, after Feb 2010. ITT CI of aGVHD was 0.32 (95% CI: 0.21, 0.43) and 0.46 (95% CI: 0.37, 0.54) prior to and after Feb 2010, respectively (p=.11). Eighty-six patients (44%) were considered at high risk for IFI pre-HSCT by standard criteria. The rate of IFI post HCT was 9/69 (13%) and 18/127 (14%) prior to and after Feb 2010, respectively. Proven (n=2) or probable (n=2) IFI was only observed in 4 patients (2%), with the remainder having possible IFI. Subgroup analysis did not find any statistical difference between the two groups for aGVHD analyzed by stem cell source or IFI analyzed by pre-HSCT risk. In addition, there was no difference in the number of dose modifications required and when patients were analyzed by actual day of azole start.

Conclusion: Delaying the start of triazole prophylaxis till 7 days after allo-HSCT does not affect outcomes, including achieving and maintaining therapeutic levels of GVHD prophylaxis, and incidence of aGVHD or IFI.