Contextual Constrained Learning for Dose-Finding Clinical Trials

Clinical trials in the medical domain are constrained by budgets. The number of patients that can be recruited is therefore limited. When a patient population is heterogeneous, this creates difficulties in learning subgroup specific responses to a particular drug and especially for a variety of dosages. In addition, patient recruitment can be difficult by the fact that clinical trials do not aim to provide a benefit to any given patient in the trial. In this paper, we propose C3T-Budget, a contextual constrained clinical trial algorithm for dose-finding under both budget and safety constraints. The algorithm aims to maximize drug efficacy within the clinical trial while also learning about the drug being tested. C3T-Budget recruits patients with consideration of the remaining budget, the remaining time, and the characteristics of each group, such as the population distribution, estimated expected efficacy, and estimation credibility. In addition, the algorithm aims to avoid unsafe dosages. These characteristics are further illustrated in a simulated clinical trial study, which corroborates the theoretical analysis and demonstrates an efficient budget usage as well as a balanced learning-treatment trade-off.

[1]  Doina Precup,et al.  Learning Modular Safe Policies in the Bandit Setting with Application to Adaptive Clinical Trials , 2019, AISafety@IJCAI.

[2]  Lisa V. Hampson,et al.  Adaptive designs in clinical trials: why use them, and how to run and report them , 2018, BMC Medicine.

[3]  Ravishankar K. Iyer,et al.  A Contextual-Bandit-Based Approach for Informed Decision-Making in Clinical Trials , 2018, Life.

[4]  Peter Auer,et al.  Finite-time Analysis of the Multiarmed Bandit Problem , 2002, Machine Learning.

[5]  S. McColley,et al.  Efficacy and safety of ivacaftor in patients with cystic fibrosis who have an Arg117His-CFTR mutation: a double-blind, randomised controlled trial. , 2015, The Lancet. Respiratory medicine.

[6]  Jay J H Park,et al.  Critical concepts in adaptive clinical trials , 2018, Clinical epidemiology.

[7]  A. Polyzos.  Activity of SU11248, a multitargeted inhibitor of vascular endothelial growth factor receptor and platelet-derived growth factor receptor, in patients with metastatic renal cell carcinoma and various other solid tumors , 2008, The Journal of Steroid Biochemistry and Molecular Biology.

[8]  Ying Yuan,et al.  Phase I/II dose-finding design for molecularly targeted agent: Plateau determination using adaptive randomization , 2018, Statistical methods in medical research.

[9]  N. Wages,et al.  A Phase I/II adaptive design for heterogeneous groups with application to a stereotactic body radiation therapy trial , 2015, Pharmaceutical statistics.

[10]  Mihaela van der Schaar,et al.  Sequential Patient Recruitment and Allocation for Adaptive Clinical Trials , 2018, AISTATS.

[11]  Karl Atkin,et al.  Why ethnic minority groups are under-represented in clinical trials: a review of the literature. , 2004, Health & social care in the community.

[12]  Li Zhou,et al.  A Survey on Contextual Multi-armed Bandits , 2015, ArXiv.

[13]  B E Storer,et al.  Design and analysis of phase I clinical trials. , 1989, Biometrics.

[14]  S. Sym,et al.  A UGT1A1 genotype-directed phase I study of irinotecan (CPT-11) combined with fixed dose of capecitabine in patients with metastatic colorectal cancer (mCRC). , 2016, Journal of Clinical Oncology.

[15]  Aurélien Garivier,et al.  The KL-UCB Algorithm for Bounded Stochastic Bandits and Beyond , 2011, COLT.

[16]  R. Srikant,et al.  Algorithms with Logarithmic or Sublinear Regret for Constrained Contextual Bandits , 2015, NIPS.

[17]  R. Munos,et al.  Best Arm Identification in Multi-Armed Bandits , 2010, COLT.

[18]  A. Jimeno,et al.  A phase I study of sorafenib and vorinostat in patients with advanced solid tumors with expanded cohorts in renal cell carcinoma and non-small cell lung cancer , 2013, Investigational New Drugs.

[19]  R. Figlin,et al.  Activity of SU11248, a multitargeted inhibitor of vascular endothelial growth factor receptor and platelet-derived growth factor receptor, in patients with metastatic renal cell carcinoma. , 2006, Journal of clinical oncology : official journal of the American Society of Clinical Oncology.

[20]  Maryam Aziz,et al.  On Multi-Armed Bandit Designs for Phase I Clinical Trials , 2019, ArXiv.

[21]  Sofía S Villar,et al.  Covariate-Adjusted Response-Adaptive Randomization for Multi-Arm Clinical Trials Using a Modified Forward Looking Gittins Index Rule , 2017, Biometrics.

[22]  Maryam Aziz,et al.  On Multi-Armed Bandit Designs for Dose-Finding Clinical Trials. , 2019 .

[23]  Jack Bowden,et al.  Multi-armed Bandit Models for the Optimal Design of Clinical Trials: Benefits and Challenges. , 2015, Statistical science : a review journal of the Institute of Mathematical Statistics.

[24]  Ying Yuan,et al.  Adaptive designs for identifying optimal biological dose for molecularly targeted agents , 2014, Clinical trials.

[25]  D. Berry.  Bayesian Statistics and the Efficiency and Ethics of Clinical Trials , 2004 .

[26]  W. R. Thompson.  ON THE LIKELIHOOD THAT ONE UNKNOWN PROBABILITY EXCEEDS ANOTHER IN VIEW OF THE EVIDENCE OF TWO SAMPLES , 1933 .

[27]  J O'Quigley,et al.  Continual reassessment method: a practical design for phase 1 clinical trials in cancer. , 1990, Biometrics.