Table 4.
Summary of progression-free survival in the overall study population and by disease extent and region
| Overall study population |
Japan |
North America |
European Union |
|||||
|---|---|---|---|---|---|---|---|---|
| Axitinib/Gem | Placebo/Gem | Axitinib/Gem | Placebo/Gem | Axitinib/Gem | Placebo/Gem | Axitinib/Gem | Placebo/Gem | |
| Overall | ||||||||
| No. of patients | 314 | 316 | 58 | 56 | 77 | 81 | 132 | 132 |
| No. of events (%) | 116 (36.9) | 125 (39.6) | 12 (20.7) | 16 (28.6) | 39 (50.6) | 38 (46.9) | 44 (33.3) | 47 (35.6) |
| Median, months | 4.4 | 4.4 | 5.8 | 5.8 | 4.1 | 3.7 | 5.7 | 4.9 |
| (95% CI) | (4.0–5.6) | (3.7–5.2) | (4.8–7.2) | (4.0–10.5) | (2.9–5.3) | (3.4–5.7) | (3.7–7.5) | (3.8–7.0) |
| Hazard ratioa,b (95% CI) | 1.006 (0.779–1.298) | 0.905 (0.416–1.968) | 1.290 (0.814–2.045) | 0.908 (0.594–1.390) | ||||
| P valuec | 0.5203 | 0.5995 | 0.8635 | 0.6707 | ||||
| Locally advanced disease | ||||||||
| No. of patients | 76 | 75 | 18 | 19 | 15 | 16 | 33 | 33 |
| No. of events (%) | 22 (28.9) | 17 (22.7) | 4 (22.2) | 2 (10.5) | 7 (46.7) | 4 (25.0) | 7 (21.2) | 9 (27.3) |
| Median, months | 5.9 | 9.1 | 5.8 | 10.5 | 7.2 | 9.0 | 7.3 | 10.4 |
| (95% CI) | (4.2–7.3) | (5.8–10.6) | (5.6–NE) | (5.8–10.5) | (1.7–9.5) | (2.0–9.0) | (4.2–9.5) | (4.1–10.4) |
| Hazard ratioa (95% CI) | 1.888 (0.978–3.645) | 4.775 (0.531–42.915) | 1.477 (0.413–5.287) | 1.384 (0.500–3.832) | ||||
| P valued | 0.9732 | 0.9382 | 0.7273 | 0.7379 | ||||
| Metastatic disease | ||||||||
| No. of patients | 238 | 241 | 40 | 37 | 62 | 65 | 99 | 99 |
| No. of events (%) | 94 (39.5) | 108 (44.8) | 8 (20.0) | 14 (37.8) | 32 (51.6) | 34 (52.3) | 37 (37.4) | 38 (38.4) |
| Median, months | 4.2 | 3.8 | 7.2 | 4.1 | 4.1 | 3.7 | 3.9 | 4.1 |
| (95% CI) | (3.7–5.4) | (3.6–4.5) | (4.8–7.2) | (3.5–7.4) | (2.4–5.3) | (3.2–5.4) | (3.5–7.5) | (3.5–5.9) |
| Hazard ratioa (95% CI) | 0.897 (0.679–1.184) | 0.629 (0.259–1.527) | 1.264 (0.770–2.073) | 0.834 (0.523–1.329) | ||||
| P valued | 0.2214 | 0.1506 | 0.1718 | 0.2206 | ||||
Gem, gemcitabine; CI, confidence interval; NE, not estimable
aAxitinib/gemcitabine vs. placebo/gemcitabine; assuming proportional hazards model, a hazard ratio <1 indicates a reduction in hazard rate in favor of axitinib/gemcitabine and a hazard ratio >1 indicates a reduction in favor of placebo/gemcitabine.
bHazard ratio stratified by extent of the disease (locally advanced vs. metastatic pancreatic cancer).
cFrom a one-sided log-rank test of treatment stratified by extent of the disease (locally advanced vs. metastatic pancreatic cancer).
dFrom a one-sided, unstratified log-rank test.