Table 2.
Models of the effect of surgical margins on local recurrence (LR) in early-stage invasive breast cancer
| Number in model | Model estimates adjusted for study-specific median follow-up time | ||||
|---|---|---|---|---|---|
| Subjects | LR | Odds of LR (Odds Ratio) | 95% CI | P-value§ [P for trend] | |
| Model 1 (median study-specific median follow-up time 6.6 years) | 28162 | 1506 | - | - | |
| Margin status | <0.001 | ||||
| Negative | 21984 | 1005 | 1.0 | - | |
| Positive/close | 6178 | 501 | 1.96 | 1.72 – 2.24 | |
| Threshold distance for negative margins† | 0.12 [0.21*] | ||||
| > 0mm | 2898 | 167 | 1.47 | 0.67 – 3.20 | |
| 1mm | 6008 | 422 | 1.0 | - | |
| 2mm | 11144 | 530 | 0.95 | 0.54– 1.67 | |
| 5mm | 8112 | 386 | 0.65 | 0.34 – 1.26 | |
| Model 2 (median study-specific median follow-up time 8.7 years) | 13081 | 753 | - | - | - |
| Margin status | <0.001 | ||||
| Negative | 9033 | 393 | 1.0 | - | |
| Close | 2407 | 176 | 1.74 | 1.42 – 2.15 | |
| Positive | 1641 | 184 | 2.44 | 1.97 – 3.03 | |
| Threshold distance for negative margins† | 0.90 [0.58] | ||||
| 1mm | 2376 | 235 | 1.0 | - | |
| 2mm | 8350 | 414 | 0.91 | 0.46 – 1.80 | |
| 5mm | 2355 | 103 | 0.77 | 0.32 – 1.87 | |
§
P reports P-value for association; P in square brackets gives P for trend and reflects whether there was statistical evidence of a decrease in the odds of LR as the threshold distance for declaring negative margins increased
†
Threshold distance for negative margins based on >0mm (5 studies), 1mm (referent; 8 studies), 2mm (12 studies), and 5mm (8 studies) in model 1; and based on 1mm (referent; 6 studies), 2mm (10 studies), and 5mm (3 studies) in model 2
*
Trend tested excluding studies using >0mm (test based on 28 studies) for model 1 – see Methods