Skip to main content

Table 1 Sampling model, transform function and example of clinical endpoints for different types of one-parameter exponential families

From: IBIS: identify biomarker-based subgroups with a Bayesian enrichment design for targeted combination therapy

f(ψ(k, j))

h(ψ(k, j))

Example of clinical endpoint

Normal(ψ(k, j), σ2 )

ψ(k, j)

Percentage change in tumor size

Bernoulli(ψ(k, j))

logit(ψ(k, j))

Tumor response rate

Exponential(ψ(k, j))

log(ψ(k, j))

Progression free survival

Poisson(ψ(k, j))

log(ψ(k, j))

Number of tumor-involved nodes

  1. f(∙), sampling model; h(∙), transform function. The nuisance parameter σ2 is assumed to be known for a normal distribution