Table 3.
Univariate and multivariable analyses of clinical characteristics in predicting intracranial radiological progression-free survival
|
Characteristics |
Univariate analysis |
Multivariable analyses |
||||
|---|---|---|---|---|---|---|
| HR | 95% CI of HR | P | HR | 95% CI of HR | P | |
| Age, years (≧60 vs. <60) |
1.07 |
0.35 to 3.32 |
0.904 |
|
|
|
| Gender (male vs. female) |
1.27 |
0.42 to 3.84 |
0.673 |
|
|
|
| Smoking history (ever vs. never) |
1.59 |
0.51 to 4.94 |
0.419 |
|
|
|
| EGFR mutation status (positive vs. negative) |
0.20 |
0.05 to 0.77 |
0.020 |
0.20 |
0.05 to 0.81 |
0.025 |
| Mutant EGFR† (exon 21 vs. exon 19) |
1.49 |
0.40 to 5.61 |
0.555 |
|
|
|
| ECOG performance status |
1.66 |
0.84 to 3.27 |
0.144 |
|
|
|
| RPA class |
|
|
0.022‡ |
|
|
0.026‡ |
| Class I (reference) |
1 |
-- |
-- |
1 |
-- |
-- |
| Class II |
3.00 |
0.37 to 24.43 |
0.306 |
2.54 |
0.30 to 21.36 |
0.390 |
| Class III |
12.67 |
1.18 to 136.31 |
0.036 |
12.26 |
1.08 to 138.65 |
0.043 |
| Primary tumor status (uncontrolled vs. controlled) |
1.46 |
0.37 to 5.80 |
0.588 |
|
|
|
| Extracranial metastases (present vs. absent) |
0.71 |
0.23 to 2.19 |
0.549 |
|
|
|
| Number of BM (>3 vs. ≦3) |
1.02 |
0.33 to 3.14 |
0.970 |
|
|
|
| Size of largest BM (mm) |
1.01 |
0.97 to 1.05 |
0.560 |
|
|
|
| Hemorrhagic BM (yes vs. no) |
0.76 |
0.23 to 2.58 |
0.664 |
|
|
|
| Total dose (>40 Gy2 vs. ≦40 Gy2) |
0.86 |
0.23 to 3.28 |
0.825 |
|
|
|
| EGFR TKI during RT (yes vs. no) |
1.32 |
0.44 to 3.97 |
0.620 |
|
|
|
| Type of EGFR TKI§ (erlotinib vs. gefitinib) |
1.39 |
0.16 to 11.98 |
0.764 |
|
|
|
| Chemotherapy during RT (yes vs. no) | 0.95 | 0.26 to 3.46 | 0.940 | |||
Abbreviations: HR hazard ratio, CI confidence interval, ECOG Eastern Cooperative Oncology Group, RPA recursive partitioning analysis, BM brain metastases, Gy2 biologically equivalent dose equal to fraction size of 2 Gy, TKI tyrosine kinase inhibitor.
† n = 30.
§ n = 19.
‡ P for linear trend.