# cases of 
DLT Seen 
True Rate 
(n=6) 
10% 
20% 
0 or 1+ 
0.885 
0.655 
2++ 
0.098 
0.246 
>2+++ 
0.016 
0.099 
of Occurrence in Population 
30% 
40% 
50% 
0.421 
0.235 
0.100 
0.324 
0.311 
0.234 
0.256 
0.455 
0.656 
+ Outcomes Leading to Escalation to Next Dose: Probability of 
(0/6) or (1/6) 
++ Outcomes Leading to Adding 3 Patients: Probability of (2/6) 
+++ Outcomes Leading to Stopping: Probability of (3/6) or (4/6) or 
(5/6) or (6/6) 
-OR- 
# cases of 
DLT Seen True Rate of Occurrence in Population 
(n=9) 
10% 
20% 
30% 
40% 
50% 
0 or 1 
n/a 
n/a 
n/a 
n/a 
n/a 
2 or 3* 
0.095 
0.220 
0.254 
0.202 
0.117 
4 or 5** 
0.003 
0.026 
0.070 
0.109 
0.117 
* Outcomes Leading to Escalation to Next Dose: Probability of [(2/6) 
and (0/3)] or [(2/6) and (1/3)] 
** Outcomes Leading to Stopping: Probability of [(2/6) and 
(2/3)] or [(2/6) and (3/3)] 
With a sample size of 6 or 9 patients per dose cohort, it would be very 
likely (i.e. 77% chance) to see 33% (or more) of patients who experience a 
DLT if the true rate of occurrence was >50%. There will be less than a 
12.5% chance of calling toxicides dose limiting if the true rate of 
occurrence is 20% or less. 
8.3.2 Combined data: Treatment Periods I and II 
Exploratory analyses will be conducted using a dichotomous 
categorizations of immunologic or virologic improvement based on CD4+, 
p24, and/or plasma viremia. With 30 patients in the genetically modified 
lymphocyte group, the proportion of patients experiencing immunologic or 
virologic improvement can be estimated with a standard deviation less than 
0 . 10 . 
[148] 
Recombinant DNA Research, Volume 19 
