A comparison of a new multinomial stopping rule with stopping rules of fleming and gehan in single arm phase II cancer clinical trials

Table 1 Thresholds by DESR to compare with rules of Fleming (n₁ = 15, n₂ = 15, power = 0.8, alpha = 0.05)

Response		Early Progression		Stage 1 Drug Rejection		Stage 2 Drug Acceptance		Power	Alpha Error	EN_nul/PES_nul	EN_alt/PES_alt
r _alt	r _nul	epd _alt	epd _nul	n _1r	n _1p	n _1r+n _2r	n _1p+n _2p
0.2	0.05	0.4	0.6	≤1/15	≥8/15	≥4/30	≤14/30	0.796	0.025	16.4/0.90	27.7/0.15
0.2	0.05	0.3	0.5	≤1/15	≥6/15	≥4/30	≤11/30	0.785	0.025	16.2/0.92	27.4/0.17

EN_nul = expected number of patients accrued if a drug meets the criteria of the null hypothesis
EN_alt = expected number of patients accrued if a drug meets the criteria of the alternate hypothesis
PES_nul = probability of stopping the trial after stage I if a drug meets the criteria of the null hypothesis
PES_alt = probability of stopping the trial after stage I if a drug meets the criteria of the alternate hypothesis

ISSN: 1471-2288