Cytomegalovirus-Seronegative Plus Leukoreduced Blood Products Are Still Superior to Only Leukoreduced Products in Preventing Transfusion-Transmitted Cytomegalovirus Infection

Abstract

Introduction Cytomegalovirus (CMV) is one of the most important pathogens causing complications after hematopoietic stem cell transplantation (HSCT). Transfusion-transmitted CMV infection (TT-CMV) in CMV-seronegative patients is a major issue. Leukoreduced (LR) blood products were unavailable before 2004. However, they have gradually become widespread, and all blood products in Japan have been LR since 2007. This study aimed to assess the incidence of TT-CMV in CMV-seronegative HSCT recipients undergoing autologous HSCT (auto-HSCT) or allogeneic HSCT (allo-HSCT) from CMV-seronegative stem cell donors, who received transfusions with “CMV-seronegative products” from CMV-seronegative donors or “random donor products” from donors not tested for anti-CMV antibodies during the periods of availability and unavailability of LR products. Methods We retrospectively evaluated 3590 CMV-seronegative recipients registered in the TRUMP database of the Japanese Society for Transplantation and Cellular Therapy from 1999 to 2018. The recipients undergoing HSCT were categorized into three periods: Period A (when LR products were unavailable), 1999-2003 (n = 136); Period B (when there were both non-LR and LR products), 2004-2006 (n = 200); and Period C (when only LR products were available), 2007-2018 (n = 3254). TT-CMV was defined by the presence of CMV antigenemia and/or CMV diseases. This study was approved by the Ethics Committee of Kobe University Hospital and was conducted in accordance with the principles of the Declaration of Helsinki. Results The median age of the patients at HSCT was 40 years (range, 0-79 years). Among auto-HSCT patients (n = 1628) transfused with “CMV-negative products”(CN group) and “random donor products” (RD group), the incidences of TT-CMV in the CN and RD groups were 0% (0/4) and 3.1% (1/32) during Period A (P = 1.0), 22.2% (2/9) and 10.6% (7/66) during Period B (P = 0.29), and 3.8% (4/105) and 5.0% (71/1412) during Period C (P = 0.81), respectively. Among the allo-HSCT patients (n = 1962), the incidences of TT-CMV in the CN and RD groups were 10.0% (3/30) and 27.1% (19/70) during Period A (P = 0.07), 17.9% (5/28) and 29.9% (29/97) during Period B (P = 0.24), and 11.6% (41/355) and 28.0% (387/1382) during Period C (P < 0.001), respectively. During Period C, 99 patients in the CN group always underwent transfusion with leukocyte reduction filters, of which 11 (11.1%) were diagnosed with TT-CMV. In contrast, 202 patients in the CN group always underwent transfusion without the filters, of which 26 (12.9%) were diagnosed with TT-CMV (P = 0.78). In the RD group, 494 patients always underwent transfusion with leukocyte reduction filters, of which 124 (25.1%) were diagnosed with TT-CMV. On the other hand, 807 patients in the RD group always underwent transfusion without the filters, of which 245 (30.4%) were diagnosed with TT-CMV (P = 0.10). Multivariate analysis revealed that random donor products (hazard ratio [HR]: 3.11, 95% confidence interval [CI]: 2.13-4.52; P < 0.001) and age ≥16 years (HR: 2.90, 95% CI: 2.06-4.09; P < 0.001) were associated with an increased risk of TT-CMV in allo-HSCT. No significant difference was observed in the incidence of CMV diseases between the CN group (3.7%; 13/355) and the RD group (3.8%; 52/1382) (P = 1.00). The major CMV diseases included CMV colitis (n = 5) and pneumonia (n = 2) in the CN group, and CMV colitis (n = 30), pneumonia (n = 9), and retinitis (n = 4) in the RD group. Interestingly, among the patients with TT-CMV during Period C, 27/41 patients (65.9%) in the CN group and 333/387 patients (86.1%) in the RD group received early intervention with anti-CMV agents (P = 0.002). The limitation included that complement-fixation test for anti-CMV antibody was used in Japan Marrow Donor Program in this cohort. Conclusion This is the largest study demonstrating that CMV-seronegative plus LR products are still superior to only LR products from random donors in terms of TT-CMV prevention among allo-HSCT CMV-seronegative recipients. However, no significant difference was observed in the incidence of CMV diseases, possibly because of early intervention with anti-CMV agents.