Skip to main content
. Author manuscript; available in PMC: 2017 Nov 30.
Published in final edited form as: Stat Med. 2016 Aug 31;35(27):4924–4936. doi: 10.1002/sim.7095

Table 6.

Selection probability and number of patients treated at each dose combination using the proposed phase I/II waterfall design, with the target dose combinations in boldface.

Sc. Drug A Drug B
Simulation results
True pr(toxicity)
True pr(efficacy)
Selection percentage No selection Number of patients
1 2 3 1 2 3
1 2 0.10 0.30 0.50 0.25 0.35 0.55 7.5 20.2 0.9 0.5 11.6 19.5 4.9
1 0.03 0.10 0.28 0.20 0.30 0.50 0.1 20.2 50.6 3.5 9.1 22.2
2 2 0.10 0.30 0.50 0.20 0.30 0.40 23.5 42.0 0.9 11.0 11.3 18.8 5.0
1 0.03 0.10 0.28 0.07 0.12 0.20 0.0 3.3 19.3 3.5 7.9 22.3
3 2 0.10 0.30 0.50 0.25 0.50 0.60 7.9 36.1 0.9 0.5 12.0 19.3 4.9
1 0.03 0.10 0.28 0.20 0.32 0.50 0.3 14.4 39.9 3.5 8.6 22.3
4 2 0.10 0.30 0.50 0.04 0.07 0.10 1.1 5.7 0.3 92.5 7.5 12.5 4.7
1 0.03 0.10 0.28 0.01 0.03 0.05 0.0 0.0 0.4 3.4 4.3 15.4
5 2 0.30 0.48 0.60 0.40 0.45 0.49 36.0 2.6 0.0 10.9 18.6 5.4 0.7
1 0.12 0.30 0.48 0.20 0.40 0.60 16.2 32.1 2.2 13.3 21.0 5.4
6 2 0.30 0.48 0.60 0.36 0.50 0.60 51.9 2.8 0.0 42.6 19.3 5.5 0.7
1 0.12 0.30 0.48 0.01 0.06 0.10 2.0 0.6 0.1 8.8 16.6 5.0
7 2 0.32 0.45 0.60 0.50 0.55 0.63 45.6 4.7 0.1 47.7 18.0 5.9 0.9
1 0.10 0.15 0.30 0.01 0.02 0.03 1.9 0.0 0.0 7.8 9.3 14.1
8 2 0.42 0.49 0.55 0.17 0.33 0.55 3.4 0.6 0.1 88.9 6.8 1.6 0.2
1 0.30 0.40 0.50 0.06 0.22 0.39 2.4 4.4 0.2 12.8 5.8 1.4

“No selection” is the percentage that the method did not select any dose-combination as the optimal dose.