CMV reactivation after allogeneic HCT and relapse risk: evidence for early protection in acute myeloid leukemia

Abstract

The association between cytomegalovirus (CMV) reactivation and relapse was evaluated in a large cohort of patients with acute myeloid leukemia (AML) (n = 761), acute lymphoblastic leukemia (ALL) (n = 322), chronic myeloid leukemia (CML) (n = 646), lymphoma (n = 254), and myelodysplastic syndrome (MDS) (n = 371) who underwent allogeneic hematopoietic cell transplantation (HCT) between 1995 and 2005. In multivariable models, CMV pp65 antigenemia was associated with a decreased risk of relapse by day 100 among patients with AML (hazard ratio [HR] = 0.56; 95% confidence interval [CI], 0.3-0.9) but not in patients with ALL, lymphoma, CML, or MDS. The effect appeared to be independent of CMV viral load, acute graft-versus-host disease, or ganciclovir-associated neutropenia. At 1 year after HCT, early CMV reactivation was associated with reduced risk of relapse in all patients, but this did not reach significance for any disease subgroup. Furthermore, CMV reactivation was associated with increased nonrelapse mortality (HR = 1.31; 95% CI, 1.1-1.6) and no difference in overall mortality (HR = 1.05; 95% CI, 0.9-1.3). This report demonstrates a modest reduction in early relapse risk after HCT associated with CMV reactivation in a large cohort of patients without a benefit in overall survival.

Figure 1

Cumulative incidence of CMV antigenemia by day 100 after HCT (n = 2566).

Figure 2

Cumulative relapse incidence in the first year after HCT by disease group. Cumulative incidence estimates with 95% CI at day 100 and 1 year for each disease group.

Figure 3

Adjusted HR and 95% CI from multivariable models evaluating CMV reactivation by day 100 as a risk factor for relapse at day 100 and 1 year after HCT. Covariates: 1, disease risk (low vs high, intermediate vs high); 2, cytogenetic risk (adverse vs intermediate and favorable vs intermediate); 3, patient race (other/unknown vs white); 4, cell source (bone marrow vs PBSC); 5, donor sex (female vs male); 6, conditioning regimen (reduced intensity vs myeloablative); 7, donor and recipient CMV serostatus (D+/R− vs D−/R−, D−/R+ vs D−/R−, or D+/R+ vs D−/R−); 8, acute GVHD (grade 3-4 vs 0-2).

Figure 4

Cumulative relapse incidence by CMV antigenemia occurring day 0 to 50 after HCT among patients surviving free of relapse to day 50. Cumulative relapse incidence estimates and 95% CI at day 100 and 1 year after HCT.

Figure 5

Results of relapse analyses for all patients stratified by pretransplant CMV serology. (A) Adjusted HR and 95% CI from multivariable models evaluating CMV antigenemia by day 100 as a risk factor for relapse at day 100 and 1 year after HCT among CMV-seropositive patients (CMV R+) and CMV-seronegative patients (CMV R−). Covariates: 1, disease risk (low, standard, or high); 2, patient race (unknown/other or white); 3, donor CMV serostatus (D− vs D+); 4, cell source (bone marrow vs PBSC); 5, patient age (0-40 years or ≥41 years); 6, conditioning regimen (myeloablative vs reduced intensity). (B) Cumulative relapse incidence at 1 year after HCT by CMV antigenemia occurring before day 50 among patients surviving relapse-free to day 50 after HCT.

Figure 6

Adjusted HR and 95% CI from multivariable models evaluating CMV antigenemia before day 100 as a risk factor for nonrelapse mortality at 1 year and overall mortality at 1 year. Covariates: 1, disease risk (low, standard, or high); 2, cytogenetic risk (low, intermediate, or high); 3, age (0-40 years vs ≥40 years); 4, cell source (bone marrow vs PBSC); 5, HLA matching (matched vs mismatched); 6, conditioning regimen (reduced intensity vs myeloablative); 7, donor and recipient CMV serostatus (D−/R−, D+/R−, D−/R+, D+/R+); 8, GVHD prophylaxis (calcineurin inhibitor alone, calcineurin + methotrexate, or calcineurin inhibitor + mycophenolate mofetil); 9, acute GVHD (grades 3-4 vs 0-2); 10, transplant year (January 1995 to November 1998, December 1998 to May 2002, or June 2002 December 2005).