Skip to main content
. 2015 Dec 10;114(2):163–170. doi: 10.1038/bjc.2015.419

Table 3. Cox proportional hazards models for PFS and OS with HFS as time-dependent covariate.

    Univariate
Multivariatea
End point Time-dependent variable Hazard ratio (95% CI) P-value Hazard ratio (95% CI) P-value
PFS HFS (yes vs no) 0.577 (0.431–0.772) 0.0002 0.558 (0.409–0.760) 0.0002
  Grade 1 HFS 0.639 (0.440–0.927) 0.0184 0.536 (0.350–0.821) 0.0042
  Grade 2 HFS 0.481 (0.321–0.720) 0.0004 0.487 (0.320–0.741) 0.0008
  Grade 3 HFS 0.641 (0.411–0.999) 0.0495 0.740 (0.466–1.178) 0.2043
OS HFS (yes vs no) 0.417 (0.273–0.637) <0.0001 0.436 (0.285–0.666) 0.0001
  Grade 1 HFS 0.568 (0.334–0.968) 0.0375 0.548 (0.322–0.934) 0.0270
  Grade 2 HFS 0.417 (0.227–0.765) 0.0048 0.448 (0.243–0.824) 0.0098
  Grade 3 HFS 0.220 (0.093–0.520) 0.0006 0.247 (0.104–0.586) 0.0015

Abbreviations: CI=confidence interval; HFS=hand–foot syndrome; OS=overall survival; PFS=progression-free survival.

Grade 1 HFS was defined as minimal skin changes or dermatitis (e.g., erythema) without pain; grade 2 was defined as skin changes (e.g., peeling, blisters, bleeding, oedema) or pain, not interfering with function; and grade 3 was defined as ulcerative dermatitis or skin changes with pain interfering with function.

a

Significant baseline factors in multivariate models: Eastern Cooperative Oncology Group performance status (0 vs 1/2) in all models; oestrogen receptor/progesterone receptor status (any positive vs other) in both OS models.