Study population

A total of 202 patients with FLT3-ITD AML (sorafenib, n=100; control, n=102) were recruited from June 20, 2015, to July 21, 2018. The median age was 35 (range: 1860) years, with 102 males and 100 females. Patient characteristics are summarized in Table 1. Baseline factors were well balanced between the two groups. With a median of 18 days after sorafenib initiation, 59 of 100 patients required dose modifications due to adverse events, including 42 dose reductions, 12 dose interruptions, and 5 discontinuations. The median follow-up was 36.8 (range, 2.567.1) months post-transplantation.

EBV-DNAemia occurred in 22 (22.0%) patients in the sorafenib group and 23 (22.5%) patients in the control group (P=0.925). Four patients developed EBV-DNAemia 1 year after allo-HSCT, including two in the sorafenib group and two in the control group. The 1-year cumulative incidence of EBV-DNAemia was 22.0% (95% CI: 14.430.6%) and 22.5% (15.031.1%) in the sorafenib and control groups (HR=0.946, 95% CI: 0.5271.698, P=0.931). The 3-year cumulative incidence of EBV-DNAemia was 24.0% (16.132.8%) and 24.5% (16.633.2%) in the two groups, respectively (HR=0.930, 95% CI: 0.5311.629, P=0.937) (Fig. 2A).

Cumulative incidences of EBV-DNAemia (A), EBV-associated disease (B), CMV-DNAemia (C), and CMV-associated disease (D) in the sorafenib and control groups

Five patients developed EBV-associated diseases in the sorafenib group including 4 EBV-post-transplant lymphoproliferative diseases (PTLD) and 1 EBV-pneumonia, and 6 patients in the control group including 3 EBV-PTLD, 2 EBV-enteritis, and 1 EBV-encephalitis. Two patients developed EBV-associated diseases 1 year after allo-HSCT, including one in the sorafenib group and one in the control group. The 1-year cumulative incidence of EBV-associated diseases was 4.0% (95% CI: 1.39.2%) and 4.9% (1.810.4%) in the sorafenib and control groups (HR=0.745, 95% CI: 0.2002.779, P=0.744). The 3-year cumulative incidence of EBV-associated diseases was 5.0% (1.810.6%) and 5.9% (2.411.6%) in the two groups, respectively (HR=0.745, 95% CI: 0.2272.448, P=0.771) (Fig. 2B).

Only one patient in the control group received EBV-CTL for EBV-PTLD, and none of the patients received DLI for EBV infections. No patients in the sorafenib group died of EBV-associated diseases, and one patient in the control group died of EBV-PTLD. The 3-year cumulative mortality of EBV-associated diseases was 0.0% (95% CI: 0.00.0%) and 1.0% (0.14.8%) in the sorafenib and control groups (HR=0.016, 95% CI: 0.0150458.5, P=0.322).

CMV-DNAemia occurred in 54 (54.0%) patients in the sorafenib group and 53 (52.0%) patients in the control group (P=0.772). Seven patients developed CMV-DNAemia 100 days after allo-HSCT, including four in the sorafenib group and three in the control group. The initial and maximum CMV loads in the sorafenib group were 2070 (range, 51233,600) copies/ml and 2750 (range, 550175,000) copies/ml, compared with 1790 (range, 55012,300) copies/ml and 3120 (range, 57070,500) copies/ml in the control group (P=0.612; P=0.882). The duration of CMV-DNAemia was 16 (range, 450) days and 17 (range, 477) days in the sorafenib and control groups (P=0.904). The 1-year cumulative incidence of CMV-DNAemia was 54.0% (95% CI: 43.763.2%) and 52.0% (41.861.2%) in the sorafenib and control groups (HR=0.974, 95% CI: 0.6671.423, P=0.911). The 3-year cumulative incidence of CMV-DNAemia was 56.0% (45.665.1%) and 52.9% (42.762.1%) in the two groups, respectively (HR=0.991, 95% CI: 0.6821.441, P=0.997) (Fig. 2C).

Up to the last follow-up, 8 patients developed CMV-associated diseases in the sorafenib group including 4 CMV-pneumonia, 3 CMV-enteritis, and 1 CMV-retinitis, and 9 patients in the control group including 6 CMV-enteritis, 2 CMV-pneumonia, and 1 CMV-encephalitis. Six patients developed CMV-associated diseases 100 days after allo-HSCT, including three in the sorafenib group and three in the control group. The 1-year cumulative incidence of CMV-associated diseases was 8.0% (95% CI: 3.714.4%) and 7.8% (3.714.1%) in the sorafenib and control groups (HR=0.949, 95% CI: 0.3562.531, P=0.984). The 3-year cumulative incidence of CMV-associated diseases was 8.0% (3.714.4%) and 8.8% (4.315.3%) in the two groups, respectively (HR=0.830, 95% CI: 0.3202.155, P=0.826) (Fig. 2D).

Seven patients (four in the sorafenib group and three in the control group) received CMV-CTL for CMV infections, and none of the patients received DLI for CMV infections. Two patients in the sorafenib group and two in the control group died of CMV-associated diseases. The 3-year cumulative mortality of CMV-associated diseases was 2.0% (95% CI: 0.46.4%) and 2.0% (0.46.3%) in the sorafenib and control groups (HR=0.955, 95% CI: 0.1346.786, P=0.980).

Univariable and multivariable analyses of the risk factors for EBV and CMV infections post-transplantation are shown in Tables 2 and 3. All patients undergoing HID/MUD transplants received ATG as GVHD prophylaxis, and none of patients undergoing MSD transplants received ATG as GVHD prophylaxis. Considering there was collinearity between transplant modality (HID/MUD vs MSD) and ATG use in the conditioning (ATG vs no ATG), we only included ATG use in the analysis of risk factors for EBV/CMV infections. On multivariate analysis, ATG use was the only risk factor for EBV-DNAemia (HR=4.408, 95% CI: 1.9679.878, P<0.001) and EBV-associated diseases (HR =3.235, 95% CI: 1.0789.711, P=0.036), respectively. ATG use (HR =2.797, 95% CI: 1.7834.387, P<0.001) and aGVHD (HR =1.641, 95% CI: 1.0672.522, P=0.024) were the risk factors for CMV-DNAemia; aGVHD (HR =3.179, 95% CI: 1.1758.601, P=0.023) was the only risk factor for CMV-associated diseases. In contrast, age, sex, EBV and CMV serological status, CR status at transplantation, sorafenib use pre-transplantation and post-transplantation, and cGVHD did not show any significant influence on the risk of EBV and CMV infections.

Immune reconstitution was similar with respect to the counts of T lymphocyte subsets (CD3+, CD3+CD4+, CD3+CD8+), B lymphocytes (CD19+), and NK cells (CD3-CD56+) at 1, 3, 6, 9, and 12 months after allo-HSCT between the sorafenib and control groups (all P >0.05) (Table 4).

At the date of statistical analysis, 142 patients survived and 60 died, of whom 21 were in the sorafenib group and 39 were in the control group. Causes of death were leukemia relapse (n=31; 7 in the sorafenib group and 24 in the control group), infections (n=18; 10 in the sorafenib group and 8 in the control group), GVHD (n=8; 3 in the sorafenib group and 5 in the control group), EBV-PTLD (n=1; control group), thrombotic microangiopathy (n=1; control group), and acute left heart failure (n=1; sorafenib group). The 3-year cumulative incidence of relapse was 13.0% (95% CI: 7.320.4%) and 34.8% (25.544.2%) in the sorafenib and control groups, respectively (HR=0.306, 95% CI: 0.1620.579, P<0.001) (Fig. 3A). The 3-year NRM was 11.1% (95% CI: 5.918.3%) and 12.7% (7.120.0%) in the two groups (HR=0.689, 95% CI: 0.3081.540, P=0.656) (Fig. 3B). The 3-year OS was 79.0% (95% CI: 69.685.8%) and 61.4% (51.170.1%; HR=0.481, 95% CI: 0.2830.818, P=0.005), LFS was 75.9% (95% CI: 66.283.1%) and 52.5% (42.261.7%; HR=0.410, 95% CI: 0.2510.670, P<0.001), and GRFS was 65.8% (95% CI: 55.674.3%) and 46.6% (36.656.0%; HR=0.531, 95% CI: 0.3450.816, P=0.003), respectively, in the sorafenib and control groups (Fig. 3CE).

Cumulative incidences of leukemia relapse (A), non-relapse mortality (B), overall survival (C), leukemia-free survival (D), and GVHD-free/relapse-free survival (E) in the sorafenib and control groups. *P < 0.05

See the original post:
Effect of sorafenib maintenance on Epstein-Barr virus and cytomegalovirus infections in patients with FLT3-ITD AML undergoing allogeneic hematopoietic...