Literature review September-December 2005

  • Published on
    06-Jul-2016

  • View
    215

  • Download
    3

Transcript

PHARMACEUTICAL STATISTICSPharmaceut. Statist. 2006; 5: 6769Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/pst.204Literature Review:SeptemberDecember 2005Simon Day1,*,y and Scott D. Patterson21Medicines and Healthcare Products Regulatory Agency, Room 13-205, Market Towers,1 Nine Elms Lane, London SW8 5NQ, UK2GlaxoSmithKline Pharmaceuticals, 1250 South Collegeville Road, Collegeville, PA19426, USAINTRODUCTIONWe bring two changes to the Literature Reviews section of thejournal, beginning with this issue. Firstly, we are introducing aseparate non-clinical review, written by non-clinical experts,headed by Professor Ludwig Hothorn. As there is rather lessnon-clinical statistical material published, that review willappear biannually, in issues 1 and 3 of the journal. Secondly,the clinical review (as we might now call it) has a slight change inauthorship and, with that, a very slight change in potentialcoverage. But we hope its usefulness will remain at a steady level.This review covers the following journals received during theperiod from the middle of September 2005 to end of December2005:* Applied Statistics, volume 54, part 5.* Biometrical Journal, volume 47, part 5.* Biometrics, volume 61, parts 2, 3.* Biometrika, volume 92, part 4.* Biostatistics, volume 6, part 4.* Clinical Trials, volume 2, part 5.* Computational Statistics and Data Analysis, volume 50,parts 13.* Drug Information Journal, volume 39, part 4.* Journal of Biopharmaceutical Statistics, volume 15, part 6.* Journal of the Royal Statistical Society, Series A, volume168, part 4.* Statistics and Probability Letters, volume 74, parts 14.* Statistics in Medicine, volume 24, parts 2023.* Statistical Methods in Medical Research, volume 14, parts 5, 6.SELECTED HIGHLIGHTS FROM THELITERATUREThe theme of Statistical Methods in Medical Research was:* Part 6: Statistics in oral health research (pp. 537602).One tutorial has appeared in Statistics in Medicine:* Gurrin LC, Scurrah KJ, Hazelton ML. Tutorial inbiostatistics: spline smoothing with linear mixed models.Statistics in Medicine 2005; 24:33613381.There is also a tutorial in Clinical Trials: this one on animportant topic of missing data. It compares various ap-proaches for imputing missing values. . . from likelihood basedmethods to simple (their term) methods:* Beunckens C, Molenberghs G, Kenward MG. Directlikelihood analysis versus simple forms of imputation formissing data in randomized clinical trials. Clinical Trials2005; 2:379386.Phase IDetermining the maximum tolerated dose is the primary goal ofphase I research, and increasingly adaptive methods are beingused in clinical trials where the dose given next is dependent onprevious responses. In oncology trials for example, if a toxicityis observed, the next patient may receive a lower dose. If nosuch event is observed, the next patient receives a higher dose.This paper reviews coherence principles relating primarily tomodied continual-reassessment methods:* Cheung Y. Coherence principles in dose-nding studies.Biometrika 2005; 92:863873.Copyright # 2006 John Wiley & Sons, Ltd.Received \60\re /teciyE-mail: simon.day@mhra.gsi.gov.uk*Correspondence to: Simon Day, Medicines and HealthcareProducts Regulatory Agency, Room 13-205, Market Towers, 1Nine Elms Lane, London SW8 5NQ, UK.Nonlinear modelling of pharmacokinetic and, increasingly,pharmacodynamic and safety repeated-measures data is becom-ing more and more common in phase I and clinicalpharmacology research. This paper describes alternatives tothe multivariate normal distribution, most often used for thispurpose:* Lindsey J, Lindsey P. Multivariate distributions withcorrelation matrices for nonlinear repeated measurements.Computational Statistics and Data Analysis 2005; 50:720732.Phase IIIn a (regulatory) trials framework, Bayesian methods certainlydo not come to prominence reasons may be quite varied.Perhaps phase II might be where some deviation from the usualmight nd a place and this is where Wang et al. review usingBayesian methods to consider the posterior probability that atreatments effect is of a given magnitude. For the true, die-hardBayesians, however, justifying such approaches on the groundsthat they control frequentist error rates must be anathema!* Wang Y-G, Leung DH-Y, Li M, Tan S-B. Bayesian designswith frequentist and Bayesian error rate considerations.Statistical Methods in Medical Research 2005; 14:445456.A completely different phase II problem is addressed by Luand colleagues who combine total response rate and rate ofcomplete response as the endpoint. They develop a two-stagedesign and give guidance on stopping rules, etc.* Lu Y, Jin H, Lamborn KR. A design of phase II cancertrials using total and complete response endpoints. Statisticsin Medicine 2005; 24:31553170.Surrogate endpointsThe following paper perhaps has little to take away and use and its mathematical content is a bit heavier than the generalreading we usually highlight but all the same is worth a quicklook for those interested in the ongoing discussion of how todene surrogate endpoints:* Baker SG, Izmirlian G, Kipnis V. Resolving paradoxesinvolving surrogate end points. Journal of the RoyalStatistical Society, Series A 2005; 168:753762.MultiplicityMultiplicity ought (perhaps) to be a problem that is designedout of a trial but if that is not done, then clear, up-front, ruleson how to handle it are essential. Whether statistics should bean art, or more rule-based might be a debatable point . . .although pre-specication may need some art, once methods/approaches are pre-specied, then inevitably some element offollowing set rules necessarily follows. This paper gives anexample of setting out rules for families of related endpoints:* Chen X, Capizzi T, Binkowitz B, Quan H, Wei L, Luo X.Decision rule based multiplicity adjustment strategy.Clinical Trials 2005; 2:394399.Procedures to adjust for multiple hypotheses are of similarimportance. This paper describes a procedure for evaluating afamily of hypotheses:* Wiens B, Dmitrienko A. The fallback procedure forevaluating a single family of hypotheses. Journal ofBiopharmaceutical Statistics 2005; 15:929942.Sample size calculation and recalculationFollowing from multiplicity (above), designing studies wheneffects are necessary to be seen on more than one endpoint isnot simple. Xiong et al. use Alzheimers disease as an exampleand show how to calculate power and sample size for anappropriate intersectionunion test.* Xiong C, Yu K, Gao F, Yan Y, Zhang Z. Power and samplesize for clinical trials when efcacy is required in multipleendpoints: application to an Alzheimers treatment trial.Clinical Trials 2005; 2:387393.When many correlated outcomes or endpoints are involved,simulation-based procedures may be useful.* Bang H, Jung S, George S. Sample size calculation forsimulation-based multiple testing procedures. Journal ofBiopharmaceutical Statistics 2005; 15:957967.Interim analyses and Data Monitoring CommitteesThere seem to be two distinct types of interim analysis: those inwhich there is (virtually) complete follow-up on all the patientswho have been recruited so far (but not all the pre-plannedpatients have been recruited); and those where all patients havebeen recruited but many of them do not have complete follow-up. The rst case is about curtailing recruitment; the second caseis about curtailing follow-up. In most cases, standard (group)sequential methods (designed for case one) are used even in casetwo. Troendle et al. look more closely at this problem. It isimportant to understand that, in this second case, differenthypotheses are being tested at different follow-up times.* Troendle JF, Liu A, Wu C, Yu KF. Sequential testing forefcacy in clinical trials with non-transient effects. Statisticsin Medicine 2005; 24:32393250.Study designDesigning studies to compare population pharmacokineticresponse is not a setting where statisticians have traditionallyCopyright # 2006 John Wiley & Sons, Ltd. Pharmaceut. Statist. 2006; 5: 6769Literature Review68been involved. The need for involvement in design to supportsuch studies is discussed in:* Narukawa M, Yafune A. A note on power and samplingschedule in population pharmacokinetic studies. DrugInformation Journal 2005; 39:353359.Data analysis issuesHandling missing data can be a complex problem depending onthe mechanism (typically, missing at random, or not). Thesepapers reviewed analysis methods when missing data arepresent:* Ali M, Talukder E. Analysis of longitudinal binary datawith missing data due to dropouts. Journal of Biopharma-ceutical Statistics 2005; 15:9931007.* Sheng X, Carrie`re K. Strategies for analysing missing itemresponse data with an application to lung cancer. Biome-trical Journal 2005; 47:605615.When data are not necessarily missing but are just sparse, itmay be of interest to test for homogeneity of treatment responseacross different strata. This paper considers the topic ofdifference in risk:* Lui K. A simple test of the homogeneity of risk difference insparse data: an application to a multi-centre study.Biometrical Journal 2005; 47:654661.Analysis of multivariate data such as that encountered inmicroarrays and imaging is a similarly complex topic with thepotential for false positives inherent to such large data sets withcorrelated responses being of most concern to drug developers.* DeCook R, Nettleton D, Foster C, Wurtele E. Identifyingdifferentially expressed genes in unreplicated multiple-treatment microarray timecourse experiments. Computa-tional Statistics and Data Analysis 2005; 50:518532.* Bowman F. Spatio-temporal modelling of localised brainactivity. Biostatistics 2005; 6:558575.Meta-analysisSome have suggested methods for combining data across trialsto assess non-inferiority rather than predening a xed margin.This paper discusses several aspects of such proposals:* Lawrence J. Some remarks about the analysis of activecontrol trials. Biometrical Journal 2005; 47:616622.PharmacovigilanceModelling of doseresponse is a complex topic, complicatedfurther when the response is an unintended side-effect. Thispaper discusses a Bayesian approach to the topic:* Johnson T, Taylor J, Ten Haken R, Eisbruch A. A Bayesianmixture model relating dose to critical organs and functionalcomplication in 3D conformal radiation therapy. Biostatistics2005; 6:615632.An issue we do not often think of when taking a pill iswhether the expiry date is in the past, present, or future.Different conditions of light, temperature, etc. can have adramatic effect on pharmaceutical products. Please check thoseexpiry dates, and to nd out more on how they are derived(post-marketing), see:* Verbon F, van den Heuvel E, Vermaat C. The cluster designfor the postmarketing surveillance program. Drug Informa-tion Journal 2005; 39:369371.Regulatory issuesA special section of the Drug Information Journal is dedicatedto the topic of ICH E14, the clinical evaluation of QT/QTcinterval prolongation and proarrythmic potential for non-antiarrythmic drugs. Several of the papers contained in thespecial section are statistical in nature:* Dmitrienko A, Sides G, Winters K, Kovacs R, Rebhun D,Bloom J, Groh W, Eisenberg P. Electrocardiogram refer-ence ranges derived from a standardised clinical trialpopulation. Drug Information Journal 2005; 39:395405.* Patterson S, Jones B, Zariffa N. Modelling and interpretingQTc prolongation in clinical pharmacology studies. DrugInformation Journal 2005; 39:437445.* Hosmane B, Locke C. A simulation study of power inthorough QT/QTc studies and a normal approximation forplanning purposes. Drug Information Journal 2005; 39:447455.MiscellaneousFinally, if you are proud to be a pharmaceutical statistician andnot just a general, medical statistician (apologies to readerswho may quite legitimately be pharmaceutical but denitely notmedical), then Grieves outgoing Presidential address to theRoyal Statistical Society will inspire you. And to readers whoare not proud to be pharmaceutical statisticians, it is still wortha read for a better understanding of how our profession ismoving:* Grieve AP. The professionalization of the shoe clerk.Journal of the Royal Statistical Society Series A 2005;168:639656.Literature Review 69Copyright # 2006 John Wiley & Sons, Ltd. Pharmaceut. Statist. 2006; 5: 6769