Bias Assessment in Progression-Free Survival Analysis Using Interval-Censored Methods - Interval-Censored-Time-to-Event Data: Methods and Applications - page 298

TABLE 10.13: Empirical Power (%) at = 5% (two-sided): HR = 0.67,

60% Event Rate, 24 Weeks Median in C

Scenario Interval Right-point Mid-point Finkelstein's Sun's

6 86.3 86.3 86.7 86.3

I 8 86.0 86.0 86.0 85.9

12 85.3 85.3 85.4 85.4

6 86.4 86.2 86.4 86.5

II 8 86.2 85.9 86.1 86.4

12 85.3 85.1 85.0 85.2

6 86.7 86.8 86.9 86.3

III 8 86.6 86.7 86.5 86.2

12 85.5 85.4 85.9 85.9

6 77.8 77.8 86.9 86.4

IV 8 72.5 72.5 86.8 86.2

12 59.9 59.9 86.0 85.5

6 92.9 92.9 86.7 86.1

V 8 94.7 94.7 86.9 85.9

12 97.1 97.1 86.1 85.2

Exactlogranktestpower86.9%.

10.4 Discussions and Recommendations

In this chapter, we compared different approaches to handle PFS data, for

example, conventional approaches and interval-censored methods, in terms of

both point estimation and hypothesis testing. The empirical performance of

these methods was evaluated by Monte Carlo simulations with sample sizes

commonly used in practice. In particular, we focused on the situations when

assessment schedules are equal (or intended to be equal) between treatment

arms, and compared scenarios, where potential bias may arise in practice