blrm_formula_saturating: Build a BLRM formula with saturating interaction term in...

In OncoBayes2: Bayesian Logistic Regression for Oncology Dose-Escalation Trials

Build a BLRM formula with saturating interaction term in logit-space

Description

blrm_formula_saturating is a convenience function for generating a formula for blrm_trial and blrm_exnex with an interaction of the form:

2 \eta \, \frac{\prod_{i=1}^N (d_i \big / d_i^*)}{1 + \prod_{i=1}^N (d_i \big / d_i^*)}

Usage

blrm_formula_saturating(
  ref_doses,
  max_interaction_level = 2,
  specific_interaction_terms = NULL
)

Arguments

ref_doses	Numeric vector of reference doses with names corresponding to drug names
max_interaction_level	Highest interaction order to consider [1 - Inf]. Default: 2
specific_interaction_terms	List of custom interaction terms to generate (e.g. list(c("drug1", "drug2"), c("drug1", "drug3"))). Default: NULL

Value

The function returns an object of class blrm_formula.

References

Widmer, L.A., Bean, A., Ohlssen, D., Weber, S., Principled Drug-Drug Interaction Terms for Bayesian Logistic Regression Models of Drug Safety in Oncology Phase I Combination Trials arXiv pre-print, 2023, \Sexpr[results=rd]{tools:::Rd_expr_doi("10.48550/arXiv.2302.11437")}

Examples

ref_doses <- c(drug_A=10, drug_B=20)

# can be used with blrm_trial
blrm_formula_saturating(ref_doses)

OncoBayes2 documentation built on July 26, 2023, 5:30 p.m.