Premium
A utility‐based Bayesian phase I–II design for immunotherapy trials with progression‐free survival end point
Author(s) -
Guo Beibei,
Park Yeonhee,
Liu Suyu
Publication year - 2019
Publication title -
journal of the royal statistical society: series c (applied statistics)
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.205
H-Index - 72
eISSN - 1467-9876
pISSN - 0035-9254
DOI - 10.1111/rssc.12288
Subject(s) - immunotherapy , clinical trial , clinical endpoint , bayesian probability , oncology , clinical study design , end point , cancer immunotherapy , medicine , endpoint determination , surrogate endpoint , computer science , medical physics , cancer , artificial intelligence , real time computing
Summary Immunotherapy has been hailed as the biggest breakthrough for treating cancer since the first development of chemotherapy. The new features of immunotherapy make the traditional clinical trial paradigm increasingly inefficient and dysfunctional. We propose a Bayesian phase I–II design for immunotherapy trials called BDFIT to find the optimal biological dose (OBD). We jointly model the toxicity outcome, progression‐free survival (PFS) and immune response. PFS and toxicity are used as the primary end points to determine the OBD, whereas the immune response is used as an ancillary end point to screen out futile doses quickly and to predict the PFS when needed. A utility function is formulated to account for the risk–benefit trade‐off and to quantify the desirability of the dose. During the trial, based on accumulating data, the estimates of the model and dose desirability are continuously updated and used to guide the dose assignment and to select the OBD. The simulation study shows that the BDFIT design has desirable operating characteristics.