Table. Log-Rank Test and Stratified Log-Rank Test or Cox Proportional Hazards Regression Model Adjusting for Site Effect to Compare PFS Between Subgroups of Patients.
| Variable, level | No. | Event | Median PFS, mo (95% CI) | 1-y PFS rate (95% CI) | P value for log-rank test | Stratified log-rank or Cox model adjusting for site effect |
|---|---|---|---|---|---|---|
| All patients | 46 | 43 | 3.7 (2.9-5.7) | 0.05 (0.01-0.19) | NA | NA |
| Platinum naive | 2 | 1 | 2.3 (1.8-NR) | NA | .04 | 0.6109 for naive vs resistant; 0.0672 for sensitive vs resistanta |
| Platinum resistant | 10 | 10 | 2.2 (1.8-NR) | NA | NA | NA |
| Platinum sensitive | 34 | 32 | 4.1 (3.6-7.8) | 0.06 (0.02-0.24) | NA | NA |
| ATM loss | 5 | 5 | 3.3 (2.3-NR) | NA | .008 | 0.021b |
| DDR-GA | 24 | 22 | 5.7 (3.6-8.8) | 0.09 (0.02-0.33) | NA | NA |
| FH | 15 | 14 | 1.9 (1.8-4.7) | NA | NA | NA |
| PH | 2 | 2 | 2.5 (2.1-NR) | NA | NA | NA |
| FH, PH, or ATM loss | 22 | 21 | 2.6 (1.9-3.9) | NA | NA | NA |
Abbreviations: DDR-GA, DNA damage repair genetic alterations; FH, family history; NA, not applicable; NR, not reached; PFS, progression-free survival; PH, personal history; Platin, oxaliplatin or cisplatin.
a
From Cox proportional hazards regression model adjusting for site effect, all patients from MD Anderson Cancer Center were platinum sensitive, stratified log-rank test was not applicable.
b
From stratified log-rank test.