1 Year

N=1693

Observation

N=1697***

Trastuzumab

1 Year

N=1702***

Disease-free survival

- No. patients with event

219 (12.9%)

127 (7.5%)

570 (33.6%)

471 (27.7%)

- No. patients without event

1474 (87.1%)

1566 (92.5%)

1127 (66.4%)

1231 (72.3%)

P-value versus observation

<0.0001

< 0.0001

Hazard ratio versus observation

0.54

0.76

Recurrence-free survival

- No. patients with event

208 (12.3%)

113 (6.7%)

506 (29.8%)

399 (23.4%)

- No. patients without event

1485 (87.7%)

1580 (93.3%)

1191 (70.2%)

1303 (76.6%)

P-value versus observation

<0.0001

<0.0001

Hazard ratio versus observation

0.51

0.73

Distant disease-free survival

- No. patients with event

184 (10.9%)

99 (5.8%)

488 (28.8%)

399 (23.4%)

- No. patients without event

1508 (89.1%)

1594 (94.6%)

1209 (71.2%)

1303 (76.6%)

P-value versus observation

<0.0001

<0.0001

Hazard ratio versus observation

0.50

0.76

Overall survival (death)

- No. patients with event

40 (2.4%)

31 (1.8%)

350 (20.6%)

278 (16.3%)

- No. patients without event

1653 (97.6%)

1662 (98.2%)

1347 (79.4%)

1424 (83.7%)

P-value versus observation

0.24

0.0005

Hazard ratio versus observation

0.75

0.76

*Co-primary endpoint of DFS of 1 year versus observation met the pre-defined statistical boundary

**Final analysis (including crossover of 52% of patients from the observation arm to trastuzumab)

*** There is a discrepancy in the overall sample size due to a small number of patients who were randomized after the cut-off date for the 12-month median follow-up analysis

The efficacy results from the interim efficacy analysis crossed the protocol pre-specified statistical boundary for the comparison of 1-year of trastuzumab versus observation. After a median follow-up of 12 months, the hazard ratio (HR) for disease free survival (DFS) was 0.54 (95% CI 0.44, 0.67) which translates into an absolute benefit, in terms of a 2-year disease-free survival rate, of 7.6 percentage points (85.8% versus 78.2%) in favour of the trastuzumab arm.

A final analysis was performed after a median follow-up of 8 years, which showed that 1-year trastuzumab treatment is associated with a 24% risk reduction compared to observation only (HR=0.76, 95% CI 0.67, 0.86). This translates into an absolute benefit in terms of an 8-year disease free survival rate of 6.4 percentage points in favour of 1-year trastuzumab treatment.

In this final analysis, extending trastuzumab treatment for a duration of two years did not show additional benefit over treatment for 1 year [DFS HR in the intent to treat (ITT) population of 2 years versus 1 year=0.99 (95% CI: 0.87, 1.13), p-value=0.90 and OS HR=0.98 (0.83, 1.15); p-value=0.78]. The rate of asymptomatic cardiac dysfunction was increased in the 2-year treatment arm (8.1% versus 4.6% in the 1-year treatment arm). More patients experienced at least one grade 3 or 4 adverse event in the 2-year treatment arm (20.4%) compared with the 1-year treatment arm (16.3%).