Table 2 Proportional hazards analysis of all-cause mortality, breast cancer specific mortality, and recurrence with time-varying adipokinesa.

From: Association of circulating leptin, adiponectin, and resistin concentrations with long-term breast cancer prognosis in a German patient cohort

N/events

All-cause mortality

Breast cancer specific mortality

Risk of recurrence

3,112/649

P

3,022/623

P

3,112/401

P

3,010/381

P

2,878/464

P

2,786/443

P

HR (95% CI)

HR (95% CI)

HR (95% CI)

HR (95% CI)

HR (95% CI)

HR (95% CI)

Basic modelb

Full modelc

Basic modelb

Full modeld

Basic modelb

Full modeld

Leptin continuous

1.04 (0.99, 1.10)

0.15

0.97 (0.92, 1.04)

0.42

1.03 (0.96, 1.10)

0.44

0.94 (0.86, 1.02)

0.14

1.07 (1.01, 1.14)

0.03

1.001 (0.93, 1.08)

0.98

Baseline quintilese (ng/ml)

 < 1.82

Reference

 

Reference

 

Reference

 

Reference

 

Reference

 

Reference

 

 < 3.47

0.94 (0.74, 1.18)

 

0.89 (0.70, 1.14)

 

1.13 (0.84, 1.51)

 

1.10 (0.80,1.50)

 

0.93 (0.70, 1.24)

 

0.89 (0.66, 1.20)

 

 < 5.98

0.87 (0.68, 1.11)

 

0.82 (0.63, 1.07)

 

0.97 (0.70, 1.32)

 

0.91 (0.65, 1.28)

 

1.24 (0.94, 1.63)

 

1.13 (0.83, 1.52)

 

 < 10.66

0.99 (0.77, 1.26)

 

0.90 (0.68, 1.19)

 

1.18 (0.86, 1.60)

 

0.99 (0.70, 1.41)

 

1.06 (0.79, 1.42)

 

0.94 (0.67, 1.32)

 

10.66 + 

1.30 (1.01, 1.65)

 

0.94 (0.70, 1.26)

 

1.25 (0.91, 1.73)

 

0.92 (0.62, 1.36)

 

1.41 (1.05, 1.89)

 

1.04 (0.72, 1.48)

 

Adiponectin continuous

0.94 (0.86, 1.03)

0.18

0.99 (0.90, 1.08)

0.81

0.95 (0.85, 1.06)

0.33

0.98 (0.88, 1.11)

0.58

1.01 (0.91, 1.12)

0.93

1.07 (0.95, 1.19)

0.26

Baseline quintilese (mg/l)

 < 14.95

Reference

 

Reference

 

Reference

 

Reference

 

Reference

 

Reference

 

 < 21.26

0.97 (0.75, 1.25)

 

1.10 (0.84, 1.43)

 

0.84 (0.60, 1.16)

 

0.92 (0.66, 1.30)

 

1.02 (0.75, 1.38)

 

1.14 (0.83, 1.57)

 

 < 28.14

0.93 (0.72, 1.20)

 

1.11 (0.85, 1.45)

 

0.88 (0.63, 1.21)

 

0.98 (0.69, 1.37)

 

0.98 (0.72, 1.34)

 

1.23 (0.89, 1.70)

 

 < 39.31

0.83 (0.64, 1.08)

 

0.99 (0.76, 1.31)

 

0.93 (0.68, 1.28)

 

1.06 (0.76, 1.48)

 

1.16 (0.87, 1.56)

 

1.29 (0.94, 1.77)

 

39.31 + 

0.97 (0.76, 1.25)

 

1.16 (0.89, 1.50)

 

0.94 (0.69, 1.29)

 

1.06 (0.77, 1.47)

 

1.02 (0.76, 1.38)

 

1.23 (0.89, 1.68)

 

Resistin continuous

1.14 (1.06, 1.23)

 < 0.01

0.98 (0.90, 1.06)

0.55

1.20 (1.09, 1.32)

 < 0.01

0.97 (0.88, 1.08)

0.79

1.07 (0.98, 1.16)

0.15

0.94 (0.85, 1.03)

0.17

Baseline quintilese (ng/ml)

 < 7.87

Reference

 

Reference

 

Reference

 

Reference

 

Reference

 

Reference

 

 < 12.57

1.01 (0.75, 1.36)

 

1.02 (0.75, 1.38)

 

1.04 (0.70, 1.55)

 

1.03 (0.68, 1.56)

 

0.98 (0.69, 1.38)

 

1.05 (0.74, 1.51)

 

 < 19.09

1.20 (0.90, 1.60)

 

1.05 (0.78, 1.41)

 

1.40 (0.96, 2.05)

 

1.17 (0.78, 1.73)

 

1.04 (0.74, 1.45)

 

0.94 (0.66, 1.35)

 

 < 31.39

1.09 (0.82, 1.46)

 

0.82 (0.61, 1.01)

 

1.17 (0.80, 1.73)

 

0.83 (0.55, 1.25)

 

1.02 (0.73, 1.43)

 

0.83 (0.58, 1.19)

 

31.39 + 

1.40 (1.07, 1.84)

 

0.93 (0.70, 1.24)

 

1.79 (1.25, 2.56)

 

1.04 (0.70, 1.53)

 

1.16 (0.84, 1.60)

 

0.90 (0.64, 1.27)

 

BMI (kg/m2)

 < 22.5

  

1.29 (0.99, 1.67)

   

1.39 (1.00, 1.93)

   

0.97 (0.72, 1.31)

 

22.5- < 25

  

Reference

   

Reference

   

Reference

 

 ≥ 25- < 30

  

1.22 (0.96, 1.54)

   

1.32 (0.97, 1.80)

   

0.96 (0.74, 1.25)

 

30 + 

  

1.27 (0.95, 1.70)

   

1.30 (0.89, 1.89)

   

1.05 (0.75, 1.45)

 
  1. aProportional hazards regression models were modified for clustered data by introducing the random laboratory batch number.
  2. bModels were adjusted for continuous age, region, and time of blood draw after diagnosis.
  3. cModels were adjusted for the other adipokines, BMI, region, age at diagnosis (continuously), time between diagnosis and first blood draw, timing of blood draw in relation to chemotherapy, tumor size, nodal status, metastasis, grading, combined estrogen/progesterone receptor status, previous tumors, diabetes at baseline, CVD at baseline comprising angina pectoris, myocardial infarction, stroke, thrombosis, and peripheral artery disease), MHT use, smoking, alcohol consumption, leisure time PA at age 50 (quintiles of MET x h/wk), mode of detection by imaging (yes/no).
  4. dModels were adjusted as in (c) but not for diabetes at baseline, CVD at baseline, smoking, and alcohol consumption. Additionally adjusted for combined Her2 receptor status/trastuzumab use, radiotherapy, tamoxifen and/or aromatase inhibitor.
  5. eQuintiles were derived from baseline measurements. FU measurements were categorized into respective baseline quintiles.
Back to article page