Fewer early bacterial and viral infections following non-myeloablative vs. myeloablative conditioning for allotransplantation

Abstract

To examine the incidence and timing of infectious complications, we reviewed 141 consecutive lymphoma patients receiving allogeneic hematopoietic stem cell transplantation using either myeloablative (MA) (n = 65) or non-myeloablative (NMA) (n = 76) conditioning. The NMA cohort was older (48.4 vs 41.4 years, P < .01), more often given unrelated umbilical cord blood as a stem cell source (43% vs 9%, P < .01), had comparable CMV seropositivity (39% vs 49%) and 39% had received a prior autologous transplant. All patients received antimicrobial prophylaxis including an extended spectrum fluoroquinolone, fluconazole, and acyclovir in addition to weekly CMV surveillance. The time to initial infection was determined for each patient to analyze potential differences in types and time to onset of infection between the cohorts. For this, patients were evaluated once in each of three microbial categories bacterial, viral, and fungal (Table 1). Fatal infections were uncommon and the incidence was similar in the MA and NMA cohorts [MA: 12 (18%); NMA: 12 (16%)]. In the peritransplant period (day 0-30), the MA cohort had 2.2-fold greater primary bacterial infections, but the risks of initial bacterial infection were similar in the early (day 31-100) and late (day 101-365) post transplant periods. Viral infections were twice as frequent in the MA cohort during the peritransplant period though similar from day 30-100. Beyond day 100, the MA cohort again had 2-fold greater primary viral infections. Fungal infections developed in approximately 10% of patients in both cohorts and the risks were similar during all three time periods. These data demonstrate a significantly greater incidence of bacterial and viral peri-transplant infections using MA conditioning though infectious mortality was similar using either conditioning. Quicker engraftment and shorter periods of neutropenia may explain in part the reduced incidence of initial peri-transplant bacterial infections in the NMA cohort, but the pathophysiology underlying the later infectious risks is uncertain. Immune reconstitution is delayed after both MA and NMA conditioning but protection against infection appears similar using either treatment approach. Future studies to correlate immune reconstitution with infections are required to identify patients at greatest risk of later infections and to design new strategies for their prevention.Tabled 1MyeloablativeNon-myeloablativeP-ValueBacterial Day 0–3049% (36–62)22% (13–32)<.01 Day 31–10047% (28–66)48% (35–62)NS Day 101–36523% (0–45)11% (0–22)NSViral Day 0–3018% (9–28)9% (3–16).08 Day 31–10031% (17–45)44% (32–56)NS Day 101–36544% (25–64)21% (7–35).05Fungal Day 0–3012% (4–21)9% (3–16)NS Day 31–10012% (3–21)13% (5–22)NS Day 101–36510% (1–20)17% (7–28)NS Open table in a new tab To examine the incidence and timing of infectious complications, we reviewed 141 consecutive lymphoma patients receiving allogeneic hematopoietic stem cell transplantation using either myeloablative (MA) (n = 65) or non-myeloablative (NMA) (n = 76) conditioning. The NMA cohort was older (48.4 vs 41.4 years, P < .01), more often given unrelated umbilical cord blood as a stem cell source (43% vs 9%, P < .01), had comparable CMV seropositivity (39% vs 49%) and 39% had received a prior autologous transplant. All patients received antimicrobial prophylaxis including an extended spectrum fluoroquinolone, fluconazole, and acyclovir in addition to weekly CMV surveillance. The time to initial infection was determined for each patient to analyze potential differences in types and time to onset of infection between the cohorts. For this, patients were evaluated once in each of three microbial categories bacterial, viral, and fungal (Table 1). Fatal infections were uncommon and the incidence was similar in the MA and NMA cohorts [MA: 12 (18%); NMA: 12 (16%)]. In the peritransplant period (day 0-30), the MA cohort had 2.2-fold greater primary bacterial infections, but the risks of initial bacterial infection were similar in the early (day 31-100) and late (day 101-365) post transplant periods. Viral infections were twice as frequent in the MA cohort during the peritransplant period though similar from day 30-100. Beyond day 100, the MA cohort again had 2-fold greater primary viral infections. Fungal infections developed in approximately 10% of patients in both cohorts and the risks were similar during all three time periods. These data demonstrate a significantly greater incidence of bacterial and viral peri-transplant infections using MA conditioning though infectious mortality was similar using either conditioning. Quicker engraftment and shorter periods of neutropenia may explain in part the reduced incidence of initial peri-transplant bacterial infections in the NMA cohort, but the pathophysiology underlying the later infectious risks is uncertain. Immune reconstitution is delayed after both MA and NMA conditioning but protection against infection appears similar using either treatment approach. Future studies to correlate immune reconstitution with infections are required to identify patients at greatest risk of later infections and to design new strategies for their prevention.