Secondary Antifungal Prophylaxis with Liposomal Amphotericin B (AmB-L) in Patients Receiving Allogeneic Hematopoietic Stem Cell Transplantation (HSCT).

Relapse of a preceding invasive fungal infection (IFI) is a remarkable risk during HSCT and the optimal secondary prophylaxis has not been defined so far. The aim of present study was to investigate whether secondary antifungal prophylaxis with AmB-L permitted HSCT without fungal recurrence. A retrospective analysis of 26 leukemic patients (n=20 AML, n=3 ALL, n=2 CML, n=1 CLL) with possible (n=16), probable (n=7) or documented IFI (n=3) prior to HSCT was conducted in 4 centers. Prior IFI was diagnosed at a median of 5 months (range 1–32 months) before HSCT. Site of IFI was pulmonary in 25 cases; one patient had positive blood culture. In the 3 proven IFI, the fungal agent was mucor (n=2) and candida (n=1). All patients received only medical treatment during their primary infection including AmB-L (3 mg/Kg/d) in 17 cases, AmB lipid complex in 5 cases and voriconazole in 4 cases; median duration of primary therapy was 31 days (9–225 days). The median time interval between start of primary antifungal treatment and HSCT was 4,5 months (range 1–22 months). Patients received an allogeneic HSCT from a matched sibling (n=8), from an haploidentical relative (n=4) or from an unrelated donor (n=14); stem cell source was bone marrow in 9 cases, peripheral blood in 15 cases and umbilical cord blood in 2 cases. Conventional regimen was given to 25 and reduced intensity to 1 case. Secondary prophylaxis consisted of AmB-L at a dose of 3 mg/Kg/d in all patients except two who received 1 mg/Kg/d and 2 mg/Kg/d respectively. The median duration of secondary prophylaxis was 33 days (6–80 days). Abnormal liver function test results possibly related to the administration of AmB-L have been observed in 21% of the patients. Seven patients (27%) showed an increase of serum creatinine greater than twice the baseline value, and 1 of these discontinued AmB-L. Four additional patients discontinued AmB-L due to progression of IFI: 2 patients received voriconazole with no response, 1 received voriconazole followed by posaconazole and died of transplant-related complications without evidence of IFI and 1 received voriconazole in combination with caspofungin with complete resolution of IFI. The effect of transplant procedure was evaluated between d+30 and d+120: 8 (31%) patients had stable disease, 10 (38%) improved and 8 (31%) patients progressed. One patient had a late pulmonary fungal reactivation (day +180) due to aspergillus fumigatus and responded to salvage therapy with voriconazole. With a median follow-up of 396 days (range 32–1852 days), the probability of survival of the 26 patients submitted to allogeneic HSCT with a previous IFI was 30%. Overall 19 patients died, 5 of leukemia relapse and 8 due to transplant-related complications without evidence of fungal infection; 6 fungus related deaths (23%) occurred. Two patients with proven mold infections died of fungal progression, 1 patient with candidemia died of bacterial sepsis. In conclusion, this series of patients with a previous history of IFI shows that secondary prophylaxis with AmB-L may be an effective measure to limit reactivation of previous IFI during HSCT even in high-risk patients conditioned with conventional myeloablative regimens.