Randomized dose-escalation design for drug combination cancer trials with immunotherapy
|
|
- Bathsheba Harvey
- 6 years ago
- Views:
Transcription
1 Randomized dose-escalation design for drug combination cancer trials with immunotherapy Pavel Mozgunov, Thomas Jaki, Xavier Paoletti Medical and Pharmaceutical Statistics Research Unit, Department of Mathematics and Statistics, Lancaster University, UK and INSERM, Institut Gustave Roussy, Villejuif, France Acknowledgement: This project has received funding from the European Union s Horizon 2020 research and innovation programme under the Marie Sklodowska-Curie grant agreement No P. Mozgunov, T.Jaki and X.Paoletti WDE-based approaches to dose-escalation 1 / 20
2 Immunotherapy in Phase I clinical trials Phase I conventional paradigm: the more the better Important exceptions: molecularly targeted agents (e.g. immunotherapy). Immune system can regulate/eliminate tumours Low toxicity profile Example: immune-checkpoint proteins blocker anti-programmed-death-receptor-1 (PD1) Pembrolizumab None of the trials reached MTD Plateau found. Same toxicity/activity probability for 2 and 10 mg/kg FDA requested to focus on a lower dose level P. Mozgunov, T.Jaki and X.Paoletti WDE-based approaches to dose-escalation 2 / 20
3 Immunotherapy in Phase I clinical trials Phase I conventional paradigm: the more the better Important exceptions: molecularly targeted agents (e.g. immunotherapy). Immune system can regulate/eliminate tumours Low toxicity profile Example: immune-checkpoint proteins blocker anti-programmed-death-receptor-1 (PD1) Pembrolizumab None of the trials reached MTD Plateau found. Same toxicity/activity probability for 2 and 10 mg/kg FDA requested to focus on a lower dose level P. Mozgunov, T.Jaki and X.Paoletti WDE-based approaches to dose-escalation 2 / 20
4 Combinations with immunotherapy Immunotherapy is enough not efficacious in cancer treatment by itself Current investigations: the added value of immune checkpoint blockers to backbone therapy the added value of a new drug to an immune checkpoint blocker. One drug is administered at full dose while the other is escalated. Objectives of the trial: To find the maximum tolerated combination (MTC) To detect clinically significant difference between the MTC and standard therapy alone (required by EMA) To detect a possible dose effect in the combination P. Mozgunov, T.Jaki and X.Paoletti WDE-based approaches to dose-escalation 3 / 20
5 Combinations with immunotherapy Immunotherapy is enough not efficacious in cancer treatment by itself Current investigations: the added value of immune checkpoint blockers to backbone therapy the added value of a new drug to an immune checkpoint blocker. One drug is administered at full dose while the other is escalated. Objectives of the trial: To find the maximum tolerated combination (MTC) To detect clinically significant difference between the MTC and standard therapy alone (required by EMA) To detect a possible dose effect in the combination P. Mozgunov, T.Jaki and X.Paoletti WDE-based approaches to dose-escalation 3 / 20
6 Combinations with immunotherapy Immunotherapy is enough not efficacious in cancer treatment by itself Current investigations: the added value of immune checkpoint blockers to backbone therapy the added value of a new drug to an immune checkpoint blocker. One drug is administered at full dose while the other is escalated. Objectives of the trial: To find the maximum tolerated combination (MTC) To detect clinically significant difference between the MTC and standard therapy alone (required by EMA) To detect a possible dose effect in the combination P. Mozgunov, T.Jaki and X.Paoletti WDE-based approaches to dose-escalation 3 / 20
7 Current approach Current design: one parameter CRM design for single agent trial Advantages: Ability to find MTC with high probability Well-known properties Disadvantages: Strong monotonicity assumption No possibility of a plateau detection Does not allow a statistical comparison of toxicities P. Mozgunov, T.Jaki and X.Paoletti WDE-based approaches to dose-escalation 4 / 20
8 Current approach Current design: one parameter CRM design for single agent trial Advantages: Ability to find MTC with high probability Well-known properties Disadvantages: Strong monotonicity assumption No possibility of a plateau detection Does not allow a statistical comparison of toxicities P. Mozgunov, T.Jaki and X.Paoletti WDE-based approaches to dose-escalation 4 / 20
9 Current approach Current design: one parameter CRM design for single agent trial Advantages: Ability to find MTC with high probability Well-known properties Disadvantages: Strong monotonicity assumption No possibility of a plateau detection Does not allow a statistical comparison of toxicities P. Mozgunov, T.Jaki and X.Paoletti WDE-based approaches to dose-escalation 4 / 20
10 Proposals Flexible model: E max model A plateau in a dose-toxicity relation Ability to model the toxicity probability on the single agent alone independently Randomization between a control and an investigation arm control is standard therapy prevents a selection bias allows statistical comparison of the toxicity ethical P. Mozgunov, T.Jaki and X.Paoletti WDE-based approaches to dose-escalation 5 / 20
11 Proposals Flexible model: E max model A plateau in a dose-toxicity relation Ability to model the toxicity probability on the single agent alone independently Randomization between a control and an investigation arm control is standard therapy prevents a selection bias allows statistical comparison of the toxicity ethical P. Mozgunov, T.Jaki and X.Paoletti WDE-based approaches to dose-escalation 5 / 20
12 Bayesian CRM Combination of A (fixed) and B: d 0 = {a, b 0 }, d 1 = {a, b 1 },..., d m = {a, b m } Model p i = ψ(d i, θ); d i is a unit-less amount of drug θ is a vector of parameters Given binary outcomes, the CRM updates the posterior f j (θ) The posterior mean (!) f j 1 (θ)l(d, y, θ) f j (θ) = f R d j 1 (u)l(d, y, u)du ˆp (j) k = E(ψ(d k, a) Y j ) = (1) R d ψ(d k, u)f j (u)du (2) Main debate: choice of model ψ(d k, a) P. Mozgunov, T.Jaki and X.Paoletti WDE-based approaches to dose-escalation 6 / 20
13 Bayesian CRM Combination of A (fixed) and B: d 0 = {a, b 0 }, d 1 = {a, b 1 },..., d m = {a, b m } Model p i = ψ(d i, θ); d i is a unit-less amount of drug θ is a vector of parameters Given binary outcomes, the CRM updates the posterior f j (θ) The posterior mean (!) f j 1 (θ)l(d, y, θ) f j (θ) = f R d j 1 (u)l(d, y, u)du ˆp (j) k = E(ψ(d k, a) Y j ) = (1) R d ψ(d k, u)f j (u)du (2) Main debate: choice of model ψ(d k, a) P. Mozgunov, T.Jaki and X.Paoletti WDE-based approaches to dose-escalation 6 / 20
14 Bayesian CRM Combination of A (fixed) and B: d 0 = {a, b 0 }, d 1 = {a, b 1 },..., d m = {a, b m } Model p i = ψ(d i, θ); d i is a unit-less amount of drug θ is a vector of parameters Given binary outcomes, the CRM updates the posterior f j (θ) The posterior mean (!) f j 1 (θ)l(d, y, θ) f j (θ) = f R d j 1 (u)l(d, y, u)du ˆp (j) k = E(ψ(d k, a) Y j ) = (1) R d ψ(d k, u)f j (u)du (2) Main debate: choice of model ψ(d k, a) P. Mozgunov, T.Jaki and X.Paoletti WDE-based approaches to dose-escalation 6 / 20
15 E max model ψ(d i, E 0, E max, λ, ED 50 ) ψ(d i, θ) = E 0 + E 0 is the probability of toxicity on the control E max + E 0 is the maximum probability of toxicity ED 50 is the combination which produces E 0 + Emax 2 λ 0 is the slope factor d λ i E max d λ i + ED λ 50 (3) Skeleton construction: ( d i = ED ˆ (0) 50 ˆp i (0) Ê (0) 0 Ê (0) max + Ê (0) 0 ˆp i (0) ) 1 ˆλ (0) By definition, ˆp 0 (0) Ê (0) 0 d i = 0 P. Mozgunov, T.Jaki and X.Paoletti WDE-based approaches to dose-escalation 7 / 20
16 E max model ψ(d i, E 0, E max, λ, ED 50 ) ψ(d i, θ) = E 0 + E 0 is the probability of toxicity on the control E max + E 0 is the maximum probability of toxicity ED 50 is the combination which produces E 0 + Emax 2 λ 0 is the slope factor d λ i E max d λ i + ED λ 50 (3) Skeleton construction: ( d i = ED ˆ (0) 50 ˆp i (0) Ê (0) 0 Ê (0) max + Ê (0) 0 ˆp i (0) ) 1 ˆλ (0) By definition, ˆp 0 (0) Ê (0) 0 d i = 0 P. Mozgunov, T.Jaki and X.Paoletti WDE-based approaches to dose-escalation 7 / 20
17 E max model ψ(d i, E 0, E max, λ, ED 50 ) ψ(d i, θ) = E 0 + E 0 is the probability of toxicity on the control E max + E 0 is the maximum probability of toxicity ED 50 is the combination which produces E 0 + Emax 2 λ 0 is the slope factor d λ i E max d λ i + ED λ 50 (3) Skeleton construction: ( d i = ED ˆ (0) 50 ˆp i (0) Ê (0) 0 Ê (0) max + Ê (0) 0 ˆp i (0) ) 1 ˆλ (0) By definition, ˆp 0 (0) Ê (0) 0 d i = 0 P. Mozgunov, T.Jaki and X.Paoletti WDE-based approaches to dose-escalation 7 / 20
18 Randomization Assignment cohort-by-cohort c = c 1 + c 2 c 1 be the number of patients assigned to the current best combination c 2 be the number of patients assigned to the control, d 0. For instance, taking c 1 = 3 and c 2 = 1, one will end up with 25% of the total sample size being assigned to the control. CRM with randomization results in the majority of patients on two combinations: control and MTC. P. Mozgunov, T.Jaki and X.Paoletti WDE-based approaches to dose-escalation 8 / 20
19 Randomization Assignment cohort-by-cohort c = c 1 + c 2 c 1 be the number of patients assigned to the current best combination c 2 be the number of patients assigned to the control, d 0. For instance, taking c 1 = 3 and c 2 = 1, one will end up with 25% of the total sample size being assigned to the control. CRM with randomization results in the majority of patients on two combinations: control and MTC. P. Mozgunov, T.Jaki and X.Paoletti WDE-based approaches to dose-escalation 8 / 20
20 Randomization Assignment cohort-by-cohort c = c 1 + c 2 c 1 be the number of patients assigned to the current best combination c 2 be the number of patients assigned to the control, d 0. For instance, taking c 1 = 3 and c 2 = 1, one will end up with 25% of the total sample size being assigned to the control. CRM with randomization results in the majority of patients on two combinations: control and MTC. P. Mozgunov, T.Jaki and X.Paoletti WDE-based approaches to dose-escalation 8 / 20
21 Simulations setting Sample size n = 48 m = 7 combinations Target probability γ = 0.25; Clinically significant difference τ = 0.05 Confidence level α = 0.9 c 1 = 3, c 2 = 1 25% on the control treatment. P. Mozgunov, T.Jaki and X.Paoletti WDE-based approaches to dose-escalation 9 / 20
22 Characteristics (i) Proportion of correct recommendations (ii) Proportion of times the clinically significant difference is found (iii) Goodness of fit measure P P (P (p MTC p control τ) > α) (4) NMSE = 1 N N j=1 n i=1 (p i ψ(d i, ˆθ (j) )) 2 n i=1 (p i ˆp opt i ) 2 (5) P. Mozgunov, T.Jaki and X.Paoletti WDE-based approaches to dose-escalation 10 / 20
23 Prior and comparators Skeleton P 0 = [0.08, 0.25, 0.35, 0.45, 0.55, 0.65, 0.70, 0.75] T Information to construct prior distributions for model parameters: (i) Control: upper bound of the 95% credibility interval is (ii) Prior MTC: upper bound of the 95% credibility interval is E 0 B(0.8, ); λ Γ(1, 1); E max E 0 U[0, 1 E 0 ]; ED 50 Γ(0.4, 0.4); P. Mozgunov, T.Jaki and X.Paoletti WDE-based approaches to dose-escalation 11 / 20
24 Comparators (P1) One-parameter power model (no randomization): (L2) Two-parameter logistic models with (R) and without randomization. ψ(d i, z) = d z i. ψ(d i, β 1, β 2 ) = exp(log(β 1) + β 2 d i ) 1 + exp(log(β 1 ) + β 2 d i ) P. Mozgunov, T.Jaki and X.Paoletti WDE-based approaches to dose-escalation 12 / 20
25 Scenarios Scenario 1 Scenario 2 Scenario 3 Toxicity Scenario 4 Scenario 5 Scenario 6 Toxicity combination combination combination P. Mozgunov, T.Jaki and X.Paoletti WDE-based approaches to dose-escalation 13 / 20
26 Results (I) d 0 d 1 d 2 d 3 d 4 d 5 d 6 d 7 TR Sc E max (R) L2(R) L P Sc E max (R) L2(R) L P P. Mozgunov, T.Jaki and X.Paoletti WDE-based approaches to dose-escalation 14 / 20
27 Results (II) d 0 d 1 d 2 d 3 d 4 d 5 d 6 d 7 TR Scenario E max (R) L2(R) L P Scenario E max (R) L2(R) L P P. Mozgunov, T.Jaki and X.Paoletti WDE-based approaches to dose-escalation 15 / 20
28 Results (III) Sc 1 Sc 2 Sc 3 Sc 4 Sc 5 Sc 6 E (R) P 74.7% 16.3% 66.7% 71.5% 67.9% 24.2% NMSE L2 (R) P 71.8% 14.4% 59.7% 64.1% 74.9% 20.9% NMSE L2 P1 P 61.5% 15.7% 50.1% 50.2% 64.7 % 18.1% NMSE P 99.9% 93.7% 99.7% 99.7% 99.4% 95.3% NMSE P. Mozgunov, T.Jaki and X.Paoletti WDE-based approaches to dose-escalation 16 / 20
29 Fitted curves Scenario 1. Emax (R) Scenario 1. Logit (R) Scenario 1. Logit Toxicity Toxicity Toxicity Scenario 4. Emax (R) Scenario 4. Logit (R) Scenario 4. Logit Toxicity Toxicity Toxicity Scenario 6. Emax (R) Scenario 6. Logit (R) Scenario 6. Logit Toxicity Toxicity Toxicity combination combination combination P. Mozgunov, T.Jaki and X.Paoletti WDE-based approaches to dose-escalation 17 / 20
30 Sensitivity analysis Prior distributions: E 0 B(0, 8, ), λ Γ(c 1, c 2 ), E max E 0 U[0, 1 E 0 ], ED 50 Γ(c 3, c 4 ). Recommendation: An informative for λ and an uninformative for ED 50 Randomization proportion Recommendation: 20%-25% on the control arm P. Mozgunov, T.Jaki and X.Paoletti WDE-based approaches to dose-escalation 18 / 20
31 Conclusions Randomization and E max model allow to identify clinically significant differences with higher probability than alternatives. The cost of randomization: a small reduction in the proportion of correct recommendations in some scenarios. The randomization helps to overcome problems with fitting This design should be considered if not only MTC identification is of interest P. Mozgunov, T.Jaki and X.Paoletti WDE-based approaches to dose-escalation 19 / 20
32 Further work Large variance of number of patients on the MTC. Further investigation on the fitting problem Phase II trials: a statistical comparison of the optimal combination and control effectivenesses P. Mozgunov, T.Jaki and X.Paoletti WDE-based approaches to dose-escalation 20 / 20
Improving a safety of the Continual Reassessment Method via a modified allocation rule
Improving a safety of the Continual Reassessment Method via a modified allocation rule Pavel Mozgunov, Thomas Jaki Medical and Pharmaceutical Statistics Research Unit, Department of Mathematics and Statistics,
More informationWeighted differential entropy based approaches to dose-escalation in clinical trials
Weighted differential entropy based approaches to dose-escalation in clinical trials Pavel Mozgunov, Thomas Jaki Medical and Pharmaceutical Statistics Research Unit, Department of Mathematics and Statistics,
More informationA weighted differential entropy based approach for dose-escalation trials
A weighted differential entropy based approach for dose-escalation trials Pavel Mozgunov, Thomas Jaki Medical and Pharmaceutical Statistics Research Unit, Department of Mathematics and Statistics, Lancaster
More informationA Bayesian decision-theoretic approach to incorporating pre-clinical information into phase I clinical trials
A Bayesian decision-theoretic approach to incorporating pre-clinical information into phase I clinical trials Haiyan Zheng, Lisa V. Hampson Medical and Pharmaceutical Statistics Research Unit Department
More informationBayesian variable selection and classification with control of predictive values
Bayesian variable selection and classification with control of predictive values Eleni Vradi 1, Thomas Jaki 2, Richardus Vonk 1, Werner Brannath 3 1 Bayer AG, Germany, 2 Lancaster University, UK, 3 University
More informationPhase I design for locating schedule-specific maximum tolerated doses
Phase I design for locating schedule-specific maximum tolerated doses Nolan A. Wages, Ph.D. University of Virginia Division of Translational Research & Applied Statistics Department of Public Health Sciences
More informationDose-finding for Multi-drug Combinations
Outline Background Methods Results Conclusions Multiple-agent Trials In trials combining more than one drug, monotonicity assumption may not hold for every dose The ordering between toxicity probabilities
More informationDose-finding for Multi-drug Combinations
September 16, 2011 Outline Background Methods Results Conclusions Multiple-agent Trials In trials combining more than one drug, monotonicity assumption may not hold for every dose The ordering between
More informationBayesian decision procedures for dose escalation - a re-analysis
Bayesian decision procedures for dose escalation - a re-analysis Maria R Thomas Queen Mary,University of London February 9 th, 2010 Overview Phase I Dose Escalation Trial Terminology Regression Models
More informationCancer phase I trial design using drug combinations when a fraction of dose limiting toxicities is attributable to one or more agents
Cancer phase I trial design using drug combinations when a fraction of dose limiting toxicities is attributable to one or more agents José L. Jiménez 1, Mourad Tighiouart 2, Mauro Gasparini 1 1 Politecnico
More informationBayesian Optimal Interval Design for Phase I Clinical Trials
Bayesian Optimal Interval Design for Phase I Clinical Trials Department of Biostatistics The University of Texas, MD Anderson Cancer Center Joint work with Suyu Liu Phase I oncology trials The goal of
More informationCOMPOSITIONAL IDEAS IN THE BAYESIAN ANALYSIS OF CATEGORICAL DATA WITH APPLICATION TO DOSE FINDING CLINICAL TRIALS
COMPOSITIONAL IDEAS IN THE BAYESIAN ANALYSIS OF CATEGORICAL DATA WITH APPLICATION TO DOSE FINDING CLINICAL TRIALS M. Gasparini and J. Eisele 2 Politecnico di Torino, Torino, Italy; mauro.gasparini@polito.it
More informationAn extrapolation framework to specify requirements for drug development in children
An framework to specify requirements for drug development in children Martin Posch joint work with Gerald Hlavin Franz König Christoph Male Peter Bauer Medical University of Vienna Clinical Trials in Small
More informationBayesian concept for combined Phase 2a/b trials
Bayesian concept for combined Phase 2a/b trials /////////// Stefan Klein 07/12/2018 Agenda Phase 2a: PoC studies Phase 2b: dose finding studies Simulation Results / Discussion 2 /// Bayer /// Bayesian
More informationPubh 8482: Sequential Analysis
Pubh 8482: Sequential Analysis Joseph S. Koopmeiners Division of Biostatistics University of Minnesota Week 10 Class Summary Last time... We began our discussion of adaptive clinical trials Specifically,
More informationBAYESIAN DOSE FINDING IN PHASE I CLINICAL TRIALS BASED ON A NEW STATISTICAL FRAMEWORK
Statistica Sinica 17(2007), 531-547 BAYESIAN DOSE FINDING IN PHASE I CLINICAL TRIALS BASED ON A NEW STATISTICAL FRAMEWORK Y. Ji, Y. Li and G. Yin The University of Texas M. D. Anderson Cancer Center Abstract:
More informationSensitivity study of dose-finding methods
to of dose-finding methods Sarah Zohar 1 John O Quigley 2 1. Inserm, UMRS 717,Biostatistic Department, Hôpital Saint-Louis, Paris, France 2. Inserm, Université Paris VI, Paris, France. NY 2009 1 / 21 to
More informationADAPTIVE EXPERIMENTAL DESIGNS. Maciej Patan and Barbara Bogacka. University of Zielona Góra, Poland and Queen Mary, University of London
ADAPTIVE EXPERIMENTAL DESIGNS FOR SIMULTANEOUS PK AND DOSE-SELECTION STUDIES IN PHASE I CLINICAL TRIALS Maciej Patan and Barbara Bogacka University of Zielona Góra, Poland and Queen Mary, University of
More informationTwo-stage Adaptive Randomization for Delayed Response in Clinical Trials
Two-stage Adaptive Randomization for Delayed Response in Clinical Trials Guosheng Yin Department of Statistics and Actuarial Science The University of Hong Kong Joint work with J. Xu PSI and RSS Journal
More informationUse of frequentist and Bayesian approaches for extrapolating from adult efficacy data to design and interpret confirmatory trials in children
Use of frequentist and Bayesian approaches for extrapolating from adult efficacy data to design and interpret confirmatory trials in children Lisa Hampson, Franz Koenig and Martin Posch Department of Mathematics
More informationEvidence synthesis for a single randomized controlled trial and observational data in small populations
Evidence synthesis for a single randomized controlled trial and observational data in small populations Steffen Unkel, Christian Röver and Tim Friede Department of Medical Statistics University Medical
More informationPubH 7470: STATISTICS FOR TRANSLATIONAL & CLINICAL RESEARCH
PubH 7470: STATISTICS FOR TRANSLATIONAL & CLINICAL RESEARCH The First Step: SAMPLE SIZE DETERMINATION THE ULTIMATE GOAL The most important, ultimate step of any of clinical research is to do draw inferences;
More informationBayesian designs of phase II oncology trials to select maximum effective dose assuming monotonic dose-response relationship
Guo and Li BMC Medical Research Methodology 2014, 14:95 RESEARCH ARTICLE Open Access Bayesian designs of phase II oncology trials to select maximum effective dose assuming monotonic dose-response relationship
More informationBinomial Model. Lecture 10: Introduction to Logistic Regression. Logistic Regression. Binomial Distribution. n independent trials
Lecture : Introduction to Logistic Regression Ani Manichaikul amanicha@jhsph.edu 2 May 27 Binomial Model n independent trials (e.g., coin tosses) p = probability of success on each trial (e.g., p =! =
More informationLecture 10: Introduction to Logistic Regression
Lecture 10: Introduction to Logistic Regression Ani Manichaikul amanicha@jhsph.edu 2 May 2007 Logistic Regression Regression for a response variable that follows a binomial distribution Recall the binomial
More informationOn the efficiency of two-stage adaptive designs
On the efficiency of two-stage adaptive designs Björn Bornkamp (Novartis Pharma AG) Based on: Dette, H., Bornkamp, B. and Bretz F. (2010): On the efficiency of adaptive designs www.statistik.tu-dortmund.de/sfb823-dp2010.html
More informationMULTIPLE-OBJECTIVE DESIGNS IN A DOSE-RESPONSE EXPERIMENT
New Developments and Applications in Experimental Design IMS Lecture Notes - Monograph Series (1998) Volume 34 MULTIPLE-OBJECTIVE DESIGNS IN A DOSE-RESPONSE EXPERIMENT BY WEI ZHU AND WENG KEE WONG 1 State
More information2015 ESMO Translational Research Fellowship Report
2015 ESMO Translational Research Fellowship Report Evaluation of c-met expression and inhibition in squamous cell carcinoma of the head and neck Petr Szturz University Hospital Brno, Czech Republic This
More informationRichard D Riley was supported by funding from a multivariate meta-analysis grant from
Bayesian bivariate meta-analysis of correlated effects: impact of the prior distributions on the between-study correlation, borrowing of strength, and joint inferences Author affiliations Danielle L Burke
More informationSample size considerations in IM assays
Sample size considerations in IM assays Lanju Zhang Senior Principal Statistician, Nonclinical Biostatistics Midwest Biopharmaceutical Statistics Workshop May 6, 0 Outline Introduction Homogeneous population
More informationIndividualized Treatment Effects with Censored Data via Nonparametric Accelerated Failure Time Models
Individualized Treatment Effects with Censored Data via Nonparametric Accelerated Failure Time Models Nicholas C. Henderson Thomas A. Louis Gary Rosner Ravi Varadhan Johns Hopkins University July 31, 2018
More informationarxiv: v1 [stat.me] 11 Feb 2014
A New Approach to Designing Phase I-II Cancer Trials for Cytotoxic Chemotherapies Jay Bartroff, Tze Leung Lai, and Balasubramanian Narasimhan arxiv:1402.2550v1 [stat.me] 11 Feb 2014 Abstract Recently there
More informationIntegrating expert s knowledge in constraint learning algorithm with time-dependent exposures
Integrating expert s knowledge in constraint learning algorithm with time-dependent exposures V. Asvatourian, S. Michiels, E. Lanoy Biostatistics and Epidemiology unit, Gustave-Roussy, Villejuif, France
More informationEstimation in Flexible Adaptive Designs
Estimation in Flexible Adaptive Designs Werner Brannath Section of Medical Statistics Core Unit for Medical Statistics and Informatics Medical University of Vienna BBS and EFSPI Scientific Seminar on Adaptive
More informationA Selection Bias Conflict and Frequentist Versus Bayesian Viewpoints
A Selection Bias Conflict and Frequentist Versus Bayesian Viewpoints Micha Mandel The Hebrew University of Jerusalem, Israel, 91905 and Yosef Rinott The Hebrew University of Jerusalem, Israel, 91905 and
More informationLecture 13: More on Binary Data
Lecture 1: More on Binary Data Link functions for Binomial models Link η = g(π) π = g 1 (η) identity π η logarithmic log π e η logistic log ( π 1 π probit Φ 1 (π) Φ(η) log-log log( log π) exp( e η ) complementary
More informationGlobal Sensitivity Analysis for Repeated Measures Studies with Informative Drop-out: A Semi-Parametric Approach
Global for Repeated Measures Studies with Informative Drop-out: A Semi-Parametric Approach Daniel Aidan McDermott Ivan Diaz Johns Hopkins University Ibrahim Turkoz Janssen Research and Development September
More informationApplied Bayesian Statistics STAT 388/488
STAT 388/488 Dr. Earvin Balderama Department of Mathematics & Statistics Loyola University Chicago August 29, 207 Course Info STAT 388/488 http://math.luc.edu/~ebalderama/bayes 2 A motivating example (See
More informationOptimal Designs for Two-Arm Randomized Phase II Clinical Trials with Multiple Constraints. Wei Jiang
Optimal Designs for Two-Arm Randomized Phase II Clinical Trials with Multiple Constraints By Wei Jiang Submitted to the graduate degree program in Biostatistics and the Graduate Faculty of the University
More informationImplementing Response-Adaptive Randomization in Multi-Armed Survival Trials
Implementing Response-Adaptive Randomization in Multi-Armed Survival Trials BASS Conference 2009 Alex Sverdlov, Bristol-Myers Squibb A.Sverdlov (B-MS) Response-Adaptive Randomization 1 / 35 Joint work
More informationCollection of Biostatistics Research Archive
Collection of Biostatistics Research Archive COBRA Preprint Series Year 2016 Paper 116 Semi-Parametrics Dose Finding Methods Matthieu Clertant John O Quigley Cedar Sinai Medical Center, Cancer Center,
More informationPubh 8482: Sequential Analysis
Pubh 8482: Sequential Analysis Joseph S. Koopmeiners Division of Biostatistics University of Minnesota Week 12 Review So far... We have discussed the role of phase III clinical trials in drug development
More informationRobust design in model-based analysis of longitudinal clinical data
Robust design in model-based analysis of longitudinal clinical data Giulia Lestini, Sebastian Ueckert, France Mentré IAME UMR 1137, INSERM, University Paris Diderot, France PODE, June 0 016 Context Optimal
More informationPrevious lecture. P-value based combination. Fixed vs random effects models. Meta vs. pooled- analysis. New random effects testing.
Previous lecture P-value based combination. Fixed vs random effects models. Meta vs. pooled- analysis. New random effects testing. Interaction Outline: Definition of interaction Additive versus multiplicative
More information(1) Introduction to Bayesian statistics
Spring, 2018 A motivating example Student 1 will write down a number and then flip a coin If the flip is heads, they will honestly tell student 2 if the number is even or odd If the flip is tails, they
More informationBAYESIAN AGGREGATION OF AVERAGE DATA: AN APPLICATION IN DRUG DEVELOPMENT
Submitted to the Annals of Applied Statistics BAYESIAN AGGREGATION OF AVERAGE DATA: AN APPLICATION IN DRUG DEVELOPMENT By Sebastian Weber Andrew Gelman, Daniel Lee Michael Betancourt, Aki Vehtari, and
More informationPersonalized Treatment Selection Based on Randomized Clinical Trials. Tianxi Cai Department of Biostatistics Harvard School of Public Health
Personalized Treatment Selection Based on Randomized Clinical Trials Tianxi Cai Department of Biostatistics Harvard School of Public Health Outline Motivation A systematic approach to separating subpopulations
More informationRobustifying Trial-Derived Treatment Rules to a Target Population
1/ 39 Robustifying Trial-Derived Treatment Rules to a Target Population Yingqi Zhao Public Health Sciences Division Fred Hutchinson Cancer Research Center Workshop on Perspectives and Analysis for Personalized
More informationActivation of STING with Synthetic Cyclic Dinucleotides and Synergy with Checkpoint Inhibition
Activation of STING with Synthetic Cyclic Dinucleotides and Synergy with Checkpoint Inhibition Sarah McWhirter, PhD Director, STING Program Aduro Biotech ICI Boston 2017 Presentation 1 Disclosures Sarah
More informationPS 203 Spring 2002 Homework One - Answer Key
PS 203 Spring 2002 Homework One - Answer Key 1. If you have a home or office computer, download and install WinBUGS. If you don t have your own computer, try running WinBUGS in the Department lab. 2. The
More informationBMA-Mod : A Bayesian Model Averaging Strategy for Determining Dose-Response Relationships in the Presence of Model Uncertainty
BMA-Mod : A Bayesian Model Averaging Strategy for Determining -Response Relationships in the Presence of Model Uncertainty A. Lawrence Gould KOL 20 October, 2017 Overview and Motivation -Response Trials
More informationUsing Data Augmentation to Facilitate Conduct of Phase I-II Clinical Trials with Delayed Outcomes
Using Data Augmentation to Facilitate Conduct of Phase I-II Clinical Trials with Delayed Outcomes Ick Hoon Jin, Suyu Liu, Peter F. Thall, and Ying Yuan October 17, 2013 Abstract A practical impediment
More informationOptimal Adaptive Designs for Dose Finding in Early Phase Clinical Trials
Optimal Adaptive Designs for Dose Finding in Early Phase Clinical Trials Alam, Muhammad Iftakhar The copyright of this thesis rests with the author and no quotation from it or information derived from
More informationCHL 5225H Advanced Statistical Methods for Clinical Trials: Multiplicity
CHL 5225H Advanced Statistical Methods for Clinical Trials: Multiplicity Prof. Kevin E. Thorpe Dept. of Public Health Sciences University of Toronto Objectives 1. Be able to distinguish among the various
More informationSample Size Determination
Sample Size Determination 018 The number of subjects in a clinical study should always be large enough to provide a reliable answer to the question(s addressed. The sample size is usually determined by
More informationLatent Variable Models for Binary Data. Suppose that for a given vector of explanatory variables x, the latent
Latent Variable Models for Binary Data Suppose that for a given vector of explanatory variables x, the latent variable, U, has a continuous cumulative distribution function F (u; x) and that the binary
More information2 >1. That is, a parallel study design will require
Cross Over Design Cross over design is commonly used in various type of research for its unique feature of accounting for within subject variability. For studies with short length of treatment time, illness
More informationAdaptive Prediction of Event Times in Clinical Trials
Adaptive Prediction of Event Times in Clinical Trials Yu Lan Southern Methodist University Advisor: Daniel F. Heitjan May 8, 2017 Yu Lan (SMU) May 8, 2017 1 / 19 Clinical Trial Prediction Event-based trials:
More informationAOAS imspdf v.2017/12/08 Prn:2018/02/05; 13:43 F:aoas1122.tex; (Aurimas) p. 1
AOAS imspdf v.2017/12/08 Prn:2018/02/05; 13:43 F:aoas1122.tex; (Aurimas) p. 1 The Annals of Applied Statistics 0, Vol. 0, No. 00, 1 22 https://doi.org/10.1214/17-aoas1122 Institute of Mathematical Statistics,
More informationA decision-theoretic phase I II design for ordinal outcomes in two cycles
Biostatistics (2016), 17, 2,pp. 304 319 doi:10.1093/biostatistics/kxv045 Advance Access publication on November 9, 2015 A decision-theoretic phase I II design for ordinal outcomes in two cycles JUHEE LEE
More informationBios 6649: Clinical Trials - Statistical Design and Monitoring
Bios 6649: Clinical Trials - Statistical Design and Monitoring Spring Semester 2015 John M. Kittelson Department of Biostatistics & Informatics Colorado School of Public Health University of Colorado Denver
More informationIncorporating published univariable associations in diagnostic and prognostic modeling
Incorporating published univariable associations in diagnostic and prognostic modeling Thomas Debray Julius Center for Health Sciences and Primary Care University Medical Center Utrecht The Netherlands
More informationBayes Factor Single Arm Time-to-event User s Guide (Version 1.0.0)
Bayes Factor Single Arm Time-to-event User s Guide (Version 1.0.0) Department of Biostatistics P. O. Box 301402, Unit 1409 The University of Texas, M. D. Anderson Cancer Center Houston, Texas 77230-1402,
More informationA simulation study for comparing testing statistics in response-adaptive randomization
RESEARCH ARTICLE Open Access A simulation study for comparing testing statistics in response-adaptive randomization Xuemin Gu 1, J Jack Lee 2* Abstract Background: Response-adaptive randomizations are
More informationAssessing the Effect of Prior Distribution Assumption on the Variance Parameters in Evaluating Bioequivalence Trials
Georgia State University ScholarWorks @ Georgia State University Mathematics Theses Department of Mathematics and Statistics 8--006 Assessing the Effect of Prior Distribution Assumption on the Variance
More informationA Bayesian Nonparametric Approach to Causal Inference for Semi-competing risks
A Bayesian Nonparametric Approach to Causal Inference for Semi-competing risks Y. Xu, D. Scharfstein, P. Mueller, M. Daniels Johns Hopkins, Johns Hopkins, UT-Austin, UF JSM 2018, Vancouver 1 What are semi-competing
More informationModeling & Simulation to Improve the Design of Clinical Trials
Modeling & Simulation to Improve the Design of Clinical Trials Gary L. Rosner Oncology Biostatistics and Bioinformatics The Sidney Kimmel Comprehensive Cancer Center at Johns Hopkins Joint work with Lorenzo
More informationApproximate analysis of covariance in trials in rare diseases, in particular rare cancers
Approximate analysis of covariance in trials in rare diseases, in particular rare cancers Stephen Senn (c) Stephen Senn 1 Acknowledgements This work is partly supported by the European Union s 7th Framework
More informationUsing Joint Utilities of the Times to Response and Toxicity to Adaptively Optimize Schedule Dose Regimes
Biometrics 69, 673 682 September 2013 DOI: 10.1111/biom.12065 Using Joint Utilities of the Times to Response and Toxicity to Adaptively Optimize Schedule Dose Regimes Peter F. Thall, 1, * Hoang Q. Nguyen,
More informationOptimizing Sedative Dose in Preterm Infants Undergoing. Treatment for Respiratory Distress Syndrome
Optimizing Sedative Dose in Preterm Infants Undergoing Treatment for Respiratory Distress Syndrome Peter F. Thall 1,, Hoang Q. Nguyen 1, Sarah Zohar 2, and Pierre Maton 3 1 Dept. of Biostatistics, M.D.
More informationBayesian Applications in Biomarker Detection. Dr. Richardus Vonk Head, Research and Clinical Sciences Statistics
Bayesian Applications in Biomarker Detection Dr. Richardus Vonk Head, Research and Clinical Sciences Statistics Disclaimer The views expressed in this presentation are the personal views of the author,
More informationOptimising Group Sequential Designs. Decision Theory, Dynamic Programming. and Optimal Stopping
: Decision Theory, Dynamic Programming and Optimal Stopping Christopher Jennison Department of Mathematical Sciences, University of Bath, UK http://people.bath.ac.uk/mascj InSPiRe Conference on Methodology
More informationMulti-level Models: Idea
Review of 140.656 Review Introduction to multi-level models The two-stage normal-normal model Two-stage linear models with random effects Three-stage linear models Two-stage logistic regression with random
More informationSequential Monitoring of Clinical Trials Session 4 - Bayesian Evaluation of Group Sequential Designs
Sequential Monitoring of Clinical Trials Session 4 - Bayesian Evaluation of Group Sequential Designs Presented August 8-10, 2012 Daniel L. Gillen Department of Statistics University of California, Irvine
More informationOverrunning in Clinical Trials: a Methodological Review
Overrunning in Clinical Trials: a Methodological Review Dario Gregori Unit of Biostatistics, Epidemiology and Public Health Department of Cardiac, Thoracic and Vascular Sciences dario.gregori@unipd.it
More informationPractice of SAS Logistic Regression on Binary Pharmacodynamic Data Problems and Solutions. Alan J Xiao, Cognigen Corporation, Buffalo NY
Practice of SAS Logistic Regression on Binary Pharmacodynamic Data Problems and Solutions Alan J Xiao, Cognigen Corporation, Buffalo NY ABSTRACT Logistic regression has been widely applied to population
More informationGeneralizing the MCPMod methodology beyond normal, independent data
Generalizing the MCPMod methodology beyond normal, independent data José Pinheiro Joint work with Frank Bretz and Björn Bornkamp Novartis AG Trends and Innovations in Clinical Trial Statistics Conference
More informationMAS361. MAS361 1 Turn Over SCHOOL OF MATHEMATICS AND STATISTICS. Medical Statistics
t r r t r r t r t s s MAS361 SCHOOL OF MATHEMATICS AND STATISTICS Medical Statistics Autumn Semester 2015 16 2 hours t s 2 r t t 1 t t r t t r s t rs t2 r t s q st s r t r t r 2 t st s rs q st s rr2 q
More informationComparison of Three Calculation Methods for a Bayesian Inference of Two Poisson Parameters
Journal of Modern Applied Statistical Methods Volume 13 Issue 1 Article 26 5-1-2014 Comparison of Three Calculation Methods for a Bayesian Inference of Two Poisson Parameters Yohei Kawasaki Tokyo University
More informationBios 6649: Clinical Trials - Statistical Design and Monitoring
Bios 6649: Clinical Trials - Statistical Design and Monitoring Spring Semester 2015 John M. Kittelson Department of Biostatistics & Informatics Colorado School of Public Health University of Colorado Denver
More informationCTP-656 Tablet Confirmed Superiority Of Pharmacokinetic Profile Relative To Kalydeco in Phase I Clinical Studies
Tablet Confirmed Superiority Of Pharmacokinetic Profile Relative To Kalydeco in Phase I Clinical Studies 39 th European Cystic Fibrosis Conference 8-11 June 2016, Basel, Switzerland 2014 Concert Pharmaceuticals,
More informationStatistical Aspects of Futility Analyses. Kevin J Carroll. nd 2013
Statistical Aspects of Futility Analyses Kevin J Carroll March Spring 222013 nd 2013 1 Contents Introduction The Problem in Statistical Terms Defining Futility Three Common Futility Rules The Maths An
More informationEstimation MLE-Pandemic data MLE-Financial crisis data Evaluating estimators. Estimation. September 24, STAT 151 Class 6 Slide 1
Estimation September 24, 2018 STAT 151 Class 6 Slide 1 Pandemic data Treatment outcome, X, from n = 100 patients in a pandemic: 1 = recovered and 0 = not recovered 1 1 1 0 0 0 1 1 1 0 0 1 0 1 0 0 1 1 1
More informationA simulation study of methods for selecting subgroup-specific doses in phase 1 trials
Received: 6 April 2016 Revised: 20 September 2016 Accepted: 4 November 2016 DOI 10.1002/pst.1797 MAIN PAPER A simulation study of methods for selecting subgroup-specific doses in phase 1 trials Satoshi
More informationBayesian course - problem set 6 (lecture 7)
Bayesian course - problem set 6 (lecture 7) Ben Lambert December 7, 2016 1 A meta-analysis of beta blocker trials Table 1 shows the results of some of the 22 trials included in a meta-analysis of clinical
More informationMultiple imputation in Cox regression when there are time-varying effects of covariates
Received: 7 January 26 Revised: 4 May 28 Accepted: 7 May 28 DOI:.2/sim.7842 RESEARCH ARTICLE Multiple imputation in Cox regression when there are time-varying effects of covariates Ruth H. Keogh Tim P.
More informationGroup sequential designs for Clinical Trials with multiple treatment arms
Group sequential designs for Clinical Trials with multiple treatment arms Susanne Urach, Martin Posch Cologne, June 26, 2015 This project has received funding from the European Union s Seventh Framework
More informationBIOS 312: Precision of Statistical Inference
and Power/Sample Size and Standard Errors BIOS 312: of Statistical Inference Chris Slaughter Department of Biostatistics, Vanderbilt University School of Medicine January 3, 2013 Outline Overview and Power/Sample
More informationBayesian Drug Disease Model with Stan Using published longitudinal data summaries in population models
Bayesian Drug Disease Model with Stan Using published longitudinal data summaries in population models S. Weber 1, B. Carpenter 2, D. Lee 2, F. Y. Bois 3, A. Gelman 2, A. Racine 1 (1) Novartis, Basel;
More informationA new strategy for meta-analysis of continuous covariates in observational studies with IPD. Willi Sauerbrei & Patrick Royston
A new strategy for meta-analysis of continuous covariates in observational studies with IPD Willi Sauerbrei & Patrick Royston Overview Motivation Continuous variables functional form Fractional polynomials
More informationEnquiry. Demonstration of Uniformity of Dosage Units using Large Sample Sizes. Proposal for a new general chapter in the European Pharmacopoeia
Enquiry Demonstration of Uniformity of Dosage Units using Large Sample Sizes Proposal for a new general chapter in the European Pharmacopoeia In order to take advantage of increased batch control offered
More informationSample size re-estimation in clinical trials. Dealing with those unknowns. Chris Jennison. University of Kyoto, January 2018
Sample Size Re-estimation in Clinical Trials: Dealing with those unknowns Christopher Jennison Department of Mathematical Sciences, University of Bath, UK http://people.bath.ac.uk/mascj University of Kyoto,
More informationA Generalized Global Rank Test for Multiple, Possibly Censored, Outcomes
A Generalized Global Rank Test for Multiple, Possibly Censored, Outcomes Ritesh Ramchandani Harvard School of Public Health August 5, 2014 Ritesh Ramchandani (HSPH) Global Rank Test for Multiple Outcomes
More informationParametric dose standardization for optimizing two-agent combinations in a phase I II trial with ordinal outcomes
Appl. Statist. (2017) Parametric dose standardization for optimizing two-agent combinations in a phase I II trial with ordinal outcomes Peter F. Thall, Hoang Q. Nguyen and Ralph G. Zinner University of
More informationarxiv: v1 [stat.me] 28 Sep 2016
A Bayesian Interval Dose-Finding Design Addressing Ockham s Razor: mtpi-2 arxiv:1609.08737v1 [stat.me] 28 Sep 2016 Wentian Guo 1, Sue-Jane Wang 2,#, Shengjie Yang 3, Suiheng Lin 1, Yuan Ji 3,4, 1 Department
More informationAuxiliary-variable-enriched Biomarker Stratified Design
Auxiliary-variable-enriched Biomarker Stratified Design Ting Wang University of North Carolina at Chapel Hill tingwang@live.unc.edu 8th May, 2017 A joint work with Xiaofei Wang, Haibo Zhou, Jianwen Cai
More informationWhat is Experimental Design?
One Factor ANOVA What is Experimental Design? A designed experiment is a test in which purposeful changes are made to the input variables (x) so that we may observe and identify the reasons for change
More informationOptimal Design for Hill Model
Optimal Design for Hill Moel DA 01 D-optimal; aaptive esign; personalize esign Russell Reeve, PhD Quintiles Durham, NC, USA Copyright 01 Quintiles Outline Motivating problem, from rheumatoi arthritis Hill
More informationMODULE -4 BAYEIAN LEARNING
MODULE -4 BAYEIAN LEARNING CONTENT Introduction Bayes theorem Bayes theorem and concept learning Maximum likelihood and Least Squared Error Hypothesis Maximum likelihood Hypotheses for predicting probabilities
More informationStatistics 135 Fall 2008 Final Exam
Name: SID: Statistics 135 Fall 2008 Final Exam Show your work. The number of points each question is worth is shown at the beginning of the question. There are 10 problems. 1. [2] The normal equations
More information