response rate of 20% in this population of patients would be 
considered worthy of further evaluation. In order to have 
less than a 10% chance of missing such a response and less 
than a 10% chance of falsely determining a true response rate 
of 5%, the following design will be employed. Fifteen 
patients will be placed on a study and evaluated. If no 
patient demonstrates response, then the trial will be closed. 
If at lest 1 patient demonstrates response, then a total of 
30 patients will be enrolled in this study. 
J. Early Stopping Rules. Permanent grade 3 or recurrent grade 
4 toxicity: If such toxicity is observed in 2 of the first 
3 patients, the study will be suspended and re-evaluated. If 
no such toxicity is observed in the first 3 patients, but 
later is seen, the study will be abandoned. 
[72] 
Recombinant DNA Research, Volume 18 
