Infectious Complications After Unrelated Donor Allogeneic Hematopoietic Cell Transplantation With Or Without Alemtuzumab Based In Vivo Lymphocyte Depletion

The addition of alemtuzumab to conditioning may decrease the incidence of graft versus host disease (GVHD) after unrelated donor (URD) allogeneic hematopoietic cell transplantation (alloHCT) but may also contribute to impaired pathogen immunity due to donor lymphocyte depletion. We identified and graded infections occurring within 6 months after URD alloHCT in 78 individuals treated either with (N=40) or without (N=38) alemtuzumab during alloHCT for hematologic cancers. Conditioning included fludarabine (120 mg/m2 total) and cyclophosphamide (4,800 mg/m2 total) over 4 days with either alemtuzumab (100 mg total days -8 to -4) followed by cyclosporine (AC) or with tacrolimus, methotrexate, and sirolimus (TMS) based GVHD prophylaxis. Both groups received a filgrastim mobilized peripheral blood graft from an 8/8 or 7/8 HLA-matched unrelated donor and received GVHD prophylaxis for 180 days. Filgrastim was administered post HCT until neutrophil recovery. Infections were graded according to Cordonnier et al. (Transplantation 2006). Only proven/probable pulmonary fungal episodes by EORT/MSG criteria were included. Infection prophylaxis during the study period included trimethoprim/sulfamethoxazole BID thrice weekly after HCT engraftment and daily acyclovir or valacyclovir. Fungal prophylaxis was fluconazole until day 100 or either micafungin or voriconazole in persons receiving 1+ week of >1 mg/kg/d prednisone or equivalent. Ceftazidime was administered to persons with neutrophil count <500/μL until recovery. Persons were monitored for CMV, adenovirus, toxoplasmosis (in seropositive persons), EBV (after 2009), and HHV-6 with at least weekly blood PCR. CMV was treated with pre-emptive therapy according to published guidelines. Blood aspergillus glactomannin and beta D-glucan assays were obtained when clinically indicated. The use of AC compared to TMS resulted in greater total infections per patient (4.4 versus 2.8, P<0.0001) but similar rates of grade 3 (most serious) infections (0.5 versus 0.6, P=0.49). CMV reactivation in recipient seropositive persons was more common in AC (24/24 AC arm versus 11/24 TMS arm, P<0.0001). The median duration of CMV viremia was 6 weeks in AC versus 2 weeks in TMS (P=0.001). CMV colitis or pneumonitis occurred in 3 persons treated with AC and in 1 person treated with TMS at a mean of 47 days post alloHCT (range 15-121). Bacterial infections per patient in AC and TMS were similar (1.6 versus 1.1, P=0.08). Clostridium difficile colitis was relatively common (10 episodes AC; 11 episodes TMS). HHV-6 infections were similar (24/40 AC versus 16/38 TMS, P=0.16). HHV-6 encephalitis developed in five individuals (3 TMS, 2 AC). The median blood titer at the time CNS involvement was identified was 15,950 copies/mL (0-40,000 copies/mL) and was not predictive on onset of encephalopathy (P=0.6). EBV infection developed in 9 persons in AC and 6 persons in TMS. Three cases of EBV lymphoproliferative disorder occurred in persons treated with AC. An additional three persons in each AC and TMS were treated pre-emptively with rituximab for EBV viremia. Adenovirus infection developed in 4 persons in each TMS and AC. One person treated with AC developed adenoviral colitis. There was one episode of invasive fungal infection in TMS (Candida krusei blood stream infection) and 4 episodes in AC (3 episodes Candida blood stream infection [2 parapsilosis, 1 albicans], one pulmonary aspergillosis). In summary, the use of AC versus TMS resulted greater total infections and CMV infections but with similar bacterial and fungal infection incidence. Weekly monitoring of HHV-6 by blood PCR was ineffective at predicting development of HHV-6 CNS disease. The rate of invasive fungal infections in this cohort was low.