Table 2a Analysis of 5-year distant metastases according to the origin of the patients and period of surgery

From: Impact of tumour size on axillary involvement and distant dissemination in breast cancer

Total ( n =18159 patients)

IGR 1954–1983 (3661)

SG 1976–1990 (6778)

SG 1991–1999 (7720)

P -values*

Geometric mean tumour diameter (mm)

22.0

18.0

16.0

<10−10/<10−10

Absolute risk of distant metastases at 5 years (95% CI)

20.8% (19.5–22.2)

15.2% (14.4–16.0)

9.5% (8.8–10.2)

<10−10/<10−10

Relative risks of metastases (95% CI)

1.42 (1.29–1.56)

1 (reference)

0.61 (0.55–0.67)

<10−10/<10−10

Probit analysis: estimated tumour volume (ml) corresponding to 50% of patients with metastases at 5 years (diameter mm)

237 ml (77 mm)

305 ml (83 mm)

985 ml (123 mm)

0.07/<10−10

Probability of metastases for a 18-mm tumour (+) (%)

16.2

14.8

9.5

 

Odds ratio (logistic regression) adjusted on continuous tumour size (95% CI)

1.11 (0.99–1.23)

1 (Reference)

0.60 (0.54–0.67)

0.07/<10−10

Relative risks (Cox's model) adjusted on continuous tumour size (95% CI)

1.09 (0.99–1.20)

1 (Reference)

0.63 (0.57–0.69)

0.07/<10−10

Relative risks (Cox's model) adjusted on tumour size category (<30 vs 30 mm) (95% CI)

1.29 (1.17–1.41)

1 (Reference)

0.61 (0.55–0.67)

10−7/<10−10

  1. IGR=Institut Gustave-Roussy; SG=Stockholm-Gotland Health Care region.
  2. *P-values: the 1st P-value refers to the IGR 1954–1983 vs SG 1976–1990 comparison, the 2nd to SG 1976–1990 vs SG 1991–1999.
  3. (+) 18 mm is the geometric mean tumour diameter in the overall population.