Clinical_Trial_Biostatistics_and_Biopharmaceutical_Applications（2015）.pdf

Clinical Trial Biostatistics and Biopharmaceutical Applications presents novel biostatistical methodologies in clinical trials as well as up-to-date biostatistical applications from the pharmaceutical industry.Each chapter is self-contained with references.Divided into five sections,the book begins with emerging issues in clinical trial design and analysis.The second section examines adaptive designs in drug devel-opment,discusses the consequences of group-sequential and adaptive designs,and illustrates group sequential design in R.The third section focuses on oncology clinical trials,covering competing risks,escalation with overdose control(EWOC)dose finding,and interval-censored time-to-event data.In the fourth section,the book describes multiple test problems with applications to adaptive designs,graphical approaches to multiple testing,the estimation of simultaneous confidence intervals for multiple comparisons,and weighted para-metric multiple testing methods.The final section discusses the statistical analysis of biomarkers from omics technologies,biomarker strategies applicable to clinical development,and the statistical evaluation of surrogate endpoints.Features Examines numerous critical issues,such as dosing,assay sensitivity,competing risks,multiplicity,and the use of biomarkers Explores major advances in adaptive clinical trial designs,including group-sequential survival trials and extensions to Bayesian adaptive dose-finding designs Explains how to implement the procedures using R and other software Discusses regulatory considerations and U.S.FDA guidelines Illustrates the models and methods with many real examples and case studies from drug development,cancer research,and more This book clarifies important issues when designing and analyzing clinical trials,in-cluding several misunderstood and unresolved challenges.It will help you choose the right method for your biostatistical application.K21709w w w.c r c p r e s s.c o mClinical Trial Biostatistics and Biopharmaceutical ApplicationsClinical Trial Biostatistics and Biopharmaceutical ApplicationsEdited byWalter R.YoungDing-Geng(Din)ChenYoung ChenStatisticsK21709_cover.indd 110/6/14 10:37 AMClinical Trial Biostatistics and Biopharmaceutical ApplicationsClinical Trial Biostatistics and Biopharmaceutical ApplicationsEdited byWalter R.YoungDing-Geng(Din)ChenCRC PressTaylor&Francis Group6000 Broken Sound Parkway NW,Suite 300Boca Raton,FL 33487-2742 2015 by Taylor&Francis Group,LLCCRC Press is an imprint of Taylor&Francis Group,an Informa businessNo claim to original U.S.Government worksVersion Date:20141021International Standard Book Number-13:978-1-4822-1219-8(eBook-PDF)This book contains information obtained from authentic and highly regarded sources.Reasonable efforts have been made to publish reliable data and information,but the author and publisher cannot assume responsibility for the validity of all materials or the consequences of their use.The authors and publishers have attempted to trace the copyright holders of all material reproduced in this publication and apologize to copyright holders if permission to publish in this form has not been obtained.If any copyright material has not been acknowledged please write and let us know so we may rectify in any future reprint.Except as permitted under U.S.Copyright Law,no part of this book may be reprinted,reproduced,transmit-ted,or utilized in any form by any electronic,mechanical,or other means,now known or hereafter invented,including photocopying,microfilming,and recording,or in any information storage or retrieval system,without written permission from the publishers.For permission to photocopy or use material electronically from this work,please access (http:/ contact the Copyright Clearance Center,Inc.(CCC),222 Rosewood Drive,Danvers,MA 01923,978-750-8400.CCC is a not-for-profit organization that provides licenses and registration for a variety of users.For organizations that have been granted a photocopy license by the CCC,a separate system of payment has been arranged.Trademark Notice:Product or corporate names may be trademarks or registered trademarks,and are used only for identification and explanation without intent to infringe.Visit the Taylor&Francis Web site athttp:/and the CRC Press Web site athttp:/To my wife,Lolita Young,whose love of Atlantic City has made me continuemychairmanshipoftheconferenceforthepast29years.Ialsothankmysons,Walter and Peter,and my daughter,Katharine,for their occasional help withthe conference.Walter YoungTo my parents and parents-in-law who value higher education and hardwork,and to my wife,Ke,my son,John D.Chen,and my daughter,Jenny K.Chen,for their love and support.This book is also dedicated to my passionfor the Deming Conference!Ding-Geng(Din)ChenContentsList of Figures.xiList of Tables.xvPreface.xixEarly History of the Deming Conference,by J.Stuart Hunter.xxiiiSome Nonstatistical Reminiscences of My 44 Years of Chairingthe Deming Conference,by Walter R.Young.xxvContributors.xxxiiiSection IEmerging Issues in Clinical Trial Designand Analysis1.Emerging Challenges of Clinical Trial Methodologies inRegulatory Applications.3H.M.James Hung and Sue-Jane Wang2.Review of Randomization Methods in Clinical Trials.41Vance W.Berger and William C.Grant3.First Dose Ranging Clinical Trial Design:More Doses?Or aWider Range?.55Guojun Yuan and Naitee Ting4.Thorough QT/QTc Clinical Trials.75Yi Tsong5.Controversial(Unresolved)Issues in Noninferiority Trials.97Brian L.WiensSection IIAdaptive Clinical Trials6.Adaptive Designs in Drug Development.117Sue-Jane Wang and H.M.James HungviiviiiContents7.Optimizing Group-Sequential Designs with Focus onAdaptability:Implications of Nonproportional Hazards inClinical Trials.137Edward Lakatos8.Group Sequential Design in R.179Keaven M.AndersonSection IIIClinical Trials in Oncology9.Issues in the Design and Analysis of OncologyClinical Trials.213Stephanie Green10.Competing Risks and Their Applications in CancerClinical Trials.247Chen Hu,James J.Dignam,and Ding-Geng(Din)Chen11.Dose Finding with Escalation with Overdose Control in CancerClinical Trials.273Andr Rogatko and Mourad Tighiouart12.Interval-Censored Time-to-Event Data and Their Applicationsin Clinical Trials.307Ling Ma,Yanqin Feng,Ding-Geng(Din)Chen,and Jianguo SunSection IVMultiple Comparisons in Clinical Trials13.Introduction to Multiple Test Problems,with Applications toAdaptive Designs.337Jeff Maca,Frank Bretz,and Willi Maurer14.Graphical Approaches to Multiple Testing.349Frank Bretz,Willi Maurer,and Jeff Maca15.Pairwise Comparisons with Binary Responses:Multiplicity-Adjusted P-Values and SimultaneousConfidence Intervals.395Bernhard Klingenberg and Faraz RahmanContentsix16.Comparative Study of Five Weighted Parametric MultipleTesting Methods for Correlated Multiple Endpoints inClinical Trials.421Changchun Xie,Xuewen Lu,and Ding-Geng(Din)ChenSection VClinical Trials in a Genomic Era17.Statistical Analysis of Biomarkers from-Omics Technologies.435Herbert Pang and Hongyu Zhao18.Understanding Therapeutic Pathways via Biomarkers andOther Uses of Biomarkers in Clinical Studies.453Michael D.Hale and Scott D.Patterson19.Statistical Evaluation of Surrogate Endpointsin Clinical Studies.497Geert Molenberghs,Ariel Alonso Abad,Wim Van der Elst,Tomasz Burzykowski,and Marc BuyseIndex.537List of Figures1.1Relationship between and as a function of RR=C/P.71.2Probability that m of four regions show a nonpositivetreatment effect where sample size allocation for the fourregions:(0.20,0.10,0.30,0.40).333.1Doseresponse results from the three studies ofosteoarthritis drug.573.2Several possible dose response curves.583.3Graphical summary of the evaluation on statistical power.683.4Graphical summary of the evaluation on false positive rate.693.5Graphical summary of the evaluation on statistical powerwith a total sample.703.6Graphical summary of the evaluation on two-sided falsepositive rate with a total sample size of 420.714.1ECG waves.764.2General time course of Moxifloxacin effect of a double-blindrandomized study.815.1Tipping point plot for the example data in Table 5.1.1035.2Gatekeeping procedure for example in Table 5.2.1106.1Impact on Type I error rate one cannot rule out when the(interim)observed data are used to update thenoninferiority margin.1266.2Select among treatment dose groups based on clinicalendpoint,cutoff D=0.1296.3Adaptive monitoring-logistics models proposed inregulatory submissions.1307.1Sketch of the survival experience of a statin arm of thePROVE-IT trial.1407.2Sketch of the survival curve of a statin arm of the PROVE-ITtrial compared with two exponential curves fitted to theearlier and later portions of the statin curve.1407.3Estimating failure probabilities from graph.1417.4Piecewise constant estimate of the failure rate function.1427.5Piecewise smoothed version of failure rate function.1427.6Constructed sketch showing the general shape of survivalcurves when there is a threshold treatment lag.1457.7Constructed sketch showing the general shape of survivalcurves when the treatment effect diminishes over time.1467.8Smoothed piecewise constant hazard ratio function derivedfrom the BHAT survival curves.147xixiiList of Figures7.9Comparing power based on exponential assumption withpower based on the actual shape of the observedsurvival curves.1487.10 Two derived hazard ratio curves from the CARE trial,oneassuming proportional hazards and the other using STOPP,assuming piecewise exponential hazards.1517.11Hazard ratio curves from CARE and BHAT assumingconstant hazards.1547.12 Time-dependent hazard ratio curves from BHAT and CAREderived from KaplanMeier plots.1547.13 Comparing the original O BrienFleming andPocock boundaries.1607.14 Comparing the original O BrienFleming boundary withthe first variant of Pocock boundary.1617.15 Comparing original O BrienFleming boundary with thefirst variant of O BrienFleming boundary.1627.16 Calculating the cumulative survival probabilitiesusing conditional survival probabilities for eachof the subintervals.1668.1Examples of(a)asymmetric and(b)symmetric groupsequential design bounds for trials with a singleinterim analysis.1868.2HwangShihDeCani spending functions with =8(efficacy)and =3(futility).1909.1Efficacytoxicity contours defining sets of equallydesirable outcomes.2179.2Association of early outcome j with final outcome ontreatment arm A:change in event rate for final outcomeafter occurrence of early event.23110.1 Schematic illustration of competing risks in the multistatemodel framework.25010.2 Cumulative incidence of failures in one-sample setting forLA-NSCLC patients in RTOG 9410.26310.3 Comparing cumulative incidences of first failures forLA-NSCLC patients in RTOG 9410.26411.1Example of the logistic tolerance distribution used to modeldosetoxicity relationship.27911.2Estimated mean length of HPD of the posterior distributionof the MTD as a function of the number of patients accruedto the trial for different target probabilities of DLT.28611.3Posterior density of the MTD when the number of treatedpatients(from bottom to top)is 1,5,10,15,20,25,30,33.28811.4Posterior density of the MTD after 33 patientshave been treated.288List of Figuresxiii11.5Recommended dose of PNU as a function of anti-SEAconcentration at both the onset and the conclusion of thephase I trial.29211.6EWOC recommended dose for selected values of thetargeted probability of DLT in two distinct trials.29711.7Standard paradigm of clinical evaluation of new cancertherapies restricts the determination of dose to the initialphase of the process.29812.1Turnbull s nonparametric estimator overlaid with intcoxestimator.32613.1Effect of number of tests on type I error rate,under independence.33813.2Comparison of the rejection regions for the Bonferroni andSimes procedures.34113.3Schematic diagram of closed testing procedure.34213.4Pictorial representation of two-stage adaptive designs.34414.1Conventions for the graphical approach.35514.2Numerical example of the graphical Holm procedure withm=2 hypotheses and=0.025.35614.3Example of a graphical multiple test procedure to illustrateAlgorithm 14.1 with=0.025 and unadjusted p-valuesp1=0.015,p2=0.001,and p3=0.1.35814.4Graphical multiple test procedure fromFigure 14.3 revisited.36214.5Graphical visualization of a viable multiple test procedurefor the diabetes case study.36414.6Test strategies for composite endpoints and theircomponents.36514.7Graphical illustration of the test strategy for noninferiorityand superiority in a combination trial withmultiple endpoints.36714.8Numerical example for the diabetes case study withunadjusted p-values p1=0.1,p2=0.001,p3=0.0001,and p4=0.005.36914.9Screenshot of the GUI from the gMCP package.37014.10 Visualization of the(a)hierarchical and(b)fallbackprocedures for m=3 hypotheses.37114.11Two extensions of the original fallback procedure for m=3hypotheses.37214.12 Graphical visualization of the parallel gatekeeperprocedure with two families F1=H1,H2?F2=H3,H4.37414.13(a)Weighting strategy to test noninferiority andsuperiority for two doses.(b)Updated weighting strategyafter rejecting H1.379xivList of Figures14.14 Graphical test procedure without memory.38514.15 Entangled graphs(G1,G2)with memory.38614.16 Graphical visualization of a k-out-of-m gatekeeperprocedure with m=4 primary hypotheses,k=3,and onesecondary hypothesis H5.38915.1One-sided critical values of maxiTi1using the exactdistribution based on the binomial probabilities,theapproach based on the asymptotic multivariate normality(MVN)of(T21,T31,T41)and the approach based on theSidak inequality when comparing three groups to acontrol,for various sample sizes n1,.,n4and truebinomial probability vector.39915.2Two-sided critical values c of max(i,j)|Tij|using the exactdistribution based on the binomial probabilities,theapproach based on the asymptotic multivariate normality(MVN)of(T21,T31,.,T43),and the approach based onthe regular Bonferroni correction for all six pairwisecomparisons among K=4 groups.40015.3Simultaneous confidence region for the case of twocomparisons to control(a)and three pairwise comparisons(b,projected into the two-dimensional space),using thefirst three dose groups of the IBS trial as data.41518.1Biochemical coverage,pharmacodynamic assessment andclinical outcome.46018.2Graphical representation of the attenuation of anendogenous agonist by an antagonist.47018.3Afew analysis possibilities for various combinations ofcontinuous and binary biomarkers and responses.48318.4Common performance characteristics of screening anddiagnostic assays.48518.5Illustration of a receiver operating characteristic(ROC)curve.48618.6Hypothetical treatment X marker predictiveness plot.48719.1Psychiatric study.Individual and mean profiles for eachscale by treatment group.502List of Tables1.1Incidence of Adjudicated Events in RENAAL Trial.261.2Treatment Effect(mmHg)of a Test Drug Relative to PlaceboBased on Mean Change from Baseline in Blood Pressure at aTa