Table 5 Multivariate Cox regression sensitivity analysis of clinical endpoints at 2 years after allogeneic HCT.

From: Fludarabine-treosulfan versus fludarabine-melphalan or busulfan-cyclophosphamide conditioning in older AML or MDS patients – A clinical trial to registry data comparison

 

Acute myeloid leukemia

 

Relapse

Non-relapse mortality

Overall survival

 

HR (95% CI)

p value

HR (95% CI)

p value

HR (95% CI)

p value

FluMel (n = 256) vs FluTreo (n = 174)

1

0.84 (0.45 – 1.57)

0.588

1

0.26 (0.12 – 0.56)

0.001

1

0.34 (0.2 – 0.57)

<0.001

BuCy (n = 503) vs FluTreo (n = 174)

1

0.67 (0.40 – 1.12)

0.127

1

0.31 (0.15 – 0.66)

0.002

1

0.48 (0.3 – 0.78)

0.003

 

Myelodysplastic syndrome

 

Relapse

Non-relapse mortality

Overall survival

 

HR (95% CI)

p value

HR (95% CI)

p value

HR (95% CI)

p value

FluMel (n = 82) vs FluTreo (n = 78)

n. a.

0.57

1

0.49 (0.24 – 1.02)

0.057

1

0.6 (0.32 – 1.14)

0.117

BuCy (n = 127) vs FluTreo (n = 78)

n. a.

0.33

1

0.46 (0.19 – 1.11)

0.085

1

0.29 (0.14 – 0.60)

0.001

  1. FluMel fludarabine/melphalan, FluTreo fludarabine/treosulfan, BuCy busulfan/cyclophosphamide.
  2. p values and hazard ratios (95% CI) were derived from Cox regression analysis (overall survival) and cause-specific Cox regression analysis (relapse and non-relapse mortality) after adjustment for patient and disease characteristics, n. a. not applicable because the required number of events for HR calculation as pre-specified by the statistical analysis plan (n = 50) was not attained.
Back to article page