Table 2.

NK cell dose remains an independent predictor of RI in a multivariate analysis

PHazard ratioCI (2.5%)CI (97.5%)
RI     
 >56 years old .069 0.477 0.22 1.06 
 Alemtuzumab T-cell depletion .018 0.424 0.21 0.87 
 Sibling transplant <.001 4.588 1.99 10.60 
 High HCT-CI <.001 4.097 1.82 9.24 
 Lymphocyte dose > median .730 1.148 0.52 2.55 
 NK dose > median <.001 0.097 0.03 0.33 
aGVHD     
 Female to male transplant .320 2.049 0.49 8.52 
 Mismatched unrelated donor (9/10) .030 7.988 1.23 51.96 
 Sibling transplant .720 0.649 0.06 7.02 
 Full-intensity conditioning .006 11.414 2.01 64.73 
 CD34 dose > median .130 0.341 0.08 1.39 
 Lymphocyte dose > median .860 1.176 0.19 7.21 
 T-cell dose > median .820 1.207 0.23 6.30 
NRM     
 >56 years old .360 1.530 0.61 3.84 
 Mismatched unrelated donor (9/10) .014 2.970 1.24 7.10 
OS     
 Mismatched unrelated donor (9/10) .012 3.449 1.32 9.02 
 Alemtuzumab T-cell depletion .127 0.560 0.27 1.18 
 High HCT-CI .008 2.525 1.27 5.03 
 NK dose > median .102 0.547 0.27 1.13 
PHazard ratioCI (2.5%)CI (97.5%)
RI     
 >56 years old .069 0.477 0.22 1.06 
 Alemtuzumab T-cell depletion .018 0.424 0.21 0.87 
 Sibling transplant <.001 4.588 1.99 10.60 
 High HCT-CI <.001 4.097 1.82 9.24 
 Lymphocyte dose > median .730 1.148 0.52 2.55 
 NK dose > median <.001 0.097 0.03 0.33 
aGVHD     
 Female to male transplant .320 2.049 0.49 8.52 
 Mismatched unrelated donor (9/10) .030 7.988 1.23 51.96 
 Sibling transplant .720 0.649 0.06 7.02 
 Full-intensity conditioning .006 11.414 2.01 64.73 
 CD34 dose > median .130 0.341 0.08 1.39 
 Lymphocyte dose > median .860 1.176 0.19 7.21 
 T-cell dose > median .820 1.207 0.23 6.30 
NRM     
 >56 years old .360 1.530 0.61 3.84 
 Mismatched unrelated donor (9/10) .014 2.970 1.24 7.10 
OS     
 Mismatched unrelated donor (9/10) .012 3.449 1.32 9.02 
 Alemtuzumab T-cell depletion .127 0.560 0.27 1.18 
 High HCT-CI .008 2.525 1.27 5.03 
 NK dose > median .102 0.547 0.27 1.13 

Patient characteristics or stem cell product cell dose with a univariate P < .2 association with a clinical outcome were included in a multivariate model to assess independent predictors in a regression model. (P < .2 in bold.) Fine and Gray regression method was used for RI, aGVHD, and NRM to take competing risk into account. Cox proportional hazard regression analysis was used to identify independent associations with OS.

CI, confidence interval.

or Create an Account

Close Modal
Close Modal