EPsProg23_binary: Expected probability of a successful program deciding between...

In Sterniii3/drugdevelopR: Utility-Based Optimal Phase II/III Drug Development Planning

Expected probability of a successful program deciding between two or three phase III trials for a binary distributed outcome

Description

The function EPsProg23_binary() calculates the expected probability of a successful program with a normally distributed outcome. This function follows a special decision rule in order to determine whether two or three phase III trials should be conducted. First, two phase III trials are performed. Depending on their success, the decision for a third phase III trial is made:

• If both trials are successful, no third phase III trial will be conducted.

• If only one of the two trials is successful and the other trial has a treatment effect that points in the same direction, a third phase III trial will be conducted with a sample size of N3 = N3(ymin), which depends on an assumed minimal clinical relevant effect (ymin). The third trial then has to be significant at level alpha

• If only one of the two trials is successful and the treatment effect of the other points in opposite direction or if none of the two trials are successful, then no third trial is performed and the drug development development program is not successful. In the utility function, this will lead to a utility of -9999.

Usage

EPsProg23_binary(
  RRgo,
  n2,
  alpha,
  beta,
  w,
  p0,
  p11,
  p12,
  in1,
  in2,
  case,
  size,
  ymin
)

Arguments

RRgo	threshold value for the go/no-go decision rule
n2	total sample size for phase II; must be even number
alpha	significance level
beta	1-beta power for calculation of sample size for phase III
w	weight for mixture prior distribution
p0	assumed true rate of control group
p11	assumed true rate of treatment group
p12	assumed true rate of treatment group
in1	amount of information for p11 in terms of sample size
in2	amount of information for p12 in terms of sample size
case	choose case: "at least 1, 2 or 3 significant trials needed for approval"
size	size category "small", "medium" or "large"
ymin	assumed minimal clinical relevant effect

Value

The output of the function EPsProg23_binary() is the expected probability of a successful program.

Examples

res <- EPsProg23_binary(RRgo = 0.8, n2 = 50,  alpha = 0.025, beta = 0.1, 
                                 w = 0.6,  p0 = 0.3, p11 =  0.3, p12 = 0.5, 
                                 in1 = 300, in2 = 600, case = 2, size = "small",
                                 ymin = 0.5)

Sterniii3/drugdevelopR documentation built on Jan. 26, 2024, 6:17 a.m.