\nWhat is the benefit of attending? \;Hear about potential strategies to handle non-proportional hazards and delayed treatment effec ts from experts in the field. \;
\n
\n
Overview
\n
Designs of clinical trials with time to event primary endpoints usually re
ly on hazards being constant over time. A major challenge in immuno-oncolo
gy is the delayed onset of benefit with such therapies and the presence of
non-proportional hazards. The impact of this needs to be accounted for in
sample size calculations\, analysis methodology and reporting. In this me
eting we will examine possible strategies to handle such features\, which
may not be fully known when the trial is initiated.
\n
\nPlease
click here to view the agenda for this meeting.
Registration
\nYou can now register for this event. Registration fees are as follow
s:
\n- Members of PSI = £\;20+VAT
\n- Non-Members of PSI =
£\;115+VAT*
\n*Please note: Non-Member rate includes members
hip for the rest of the 2021 calendar year.
\nTo register for th
e \;session\, please \;
click here.
Speaker details
\n| \n Speaker \n | \n \n < strong>Biography \n | \n \n Abstract \n | \n
| \n <
img src="https://psiweb.org/images/default-source/default-album/jonathan-b
artlett-cropped.tmb-small.png?Culture=en&sfvrsn=f94fdadb_1&sf_site_temp=tr
ue&sf_site=00000000-0000-0000-0000-000000000000" data-displaymode="Thumbna
il" alt="Jonathan Bartlett cropped" title="Jonathan Bartlett cropped" />\n Jonathan Bartlett\, | \n \n
Jonathan Bartlett is a Reader in Statistics at the University of Ba th\, UK. He worked previously at AstraZeneca&rsquo\;s Statistical Innovati on Group\, and the London School of Hygiene &\; Tropical Medicine. His research interests include statistical methods for missing data &\; est imands\, covariate adjustment\, survival analysis\, and measurement error. He maintains a statistics blog at thestatsgeek.com \n |
\n \n Non-proportional h azards &ndash\; an introduction to their possible causes and interpretatio n. \nIn this introductory talk I will being by reviewing Cox&rsquo\;s proportional hazards model and the meaning of &lsqu o\;proportional hazards&rsquo\;. I will illustrate some of the different w ays in which non-proportional hazards may arise\, and in so doing\, demons trate that interpretation of time-changing hazard ratios is complicated by the fact that the survivors to a particular follow-up time in the two tre atment groups will generally systematically differ in respect of baseline prognostic variables. \n | \n
| \n
| \n
\n Kaspar Rufibach is a member of Roche's M ethods\, Collaboration\, and Outreach group and located in Basel. He does methodological research\, provides consulting to Roche statisticians and b roader project teams\, gives biostatistics trainings for statisticians and non-statisticians in- and externally\, mentors students\, and interacts w ith external partners in industry\, regulators\, and the academic communit y in various working groups and collaborations. He has co-founded and co-l eads the European special interest group &ldquo\;Estimands in oncology&rdq uo\; (sponsored by PSI and EFSPI) that currently has more than 35 members from 22 companies and several Health Authorities and works on various topi cs around estimands in oncology. Kaspar&rsquo\;s research interests are me thods to optimize study designs\, advanced survival analysis\, probability of success\, estimands and causal inference\, estimation of treatment eff ects in subgroups\, and general nonparametric statistics. Before joining R oche\, Kaspar received training and worked as a statistician at the Univer sities of Bern\, Stanford\, and Zurich. \n | \n
\n Planning a Phase 3 trial with time-to-event endpoint\, a cure proportion\, and a futility interim analys is using response. \nWith median overall surviv al (OS) of about six months and no approved drug for more than forty years \, the unmet medical need in acute myeloid leukemia is dramatic. Idasanutl in is a MDM2 antagonist that can effectively displace p53 from MDM2 to res tore p53 function\, leading to cell cycle arrest and apoptosis of cancer c ells. Planning the Phase 3 trial MIRROS comparing Idasanutlin + standard o f care against the standard of care presented with the following challenge s: (1) To survive AML a patients needs to become eligible for a bone marro w transplant\, trough achieving a complete response (CR) after induction t herapy. Planning the trial using overall survival as primary endpoint thus needs to account for a cure proportion in both\, the treatment and contro l arm. (2) MIRROS was planned based on Phase 1 data only. To mitigate the risk of directly moving to Phase 3\, a futility interim was built in the d esign\, using gates on the odds ratio for. The interim analysis was built- in the design using a mechanistic simulation model\, making assumptions on response proportions\, proportion of transplant survivors\, and OS in the se various groups. The talk describes the design in detail\, discusses sam ple size planning\, operating characteristics of the futility interim anal ysis\, and will touch upon how we plan to report the results. We conclude with sharing feedback from US and European Health Authorities on the desig n. \n | \n
| \n
| \n \n
With a PhD in Statistics from Politecnico di Torino\, Jose participa ted as an Early Stage Researcher in the Marie Curie network &ldquo\;IDEAS& rdquo\;\, where he primarily worked on Bayesian dose finding methods and n on-proportional hazards. He is currently employed by Novartis in Basel. \n | \n \n The presence of delayed effects causes a change in t he hazard ratio while the trial is ongoing since at the beginning we do no t observe any difference between treatment arms\, and after some unknown t ime point\, the differences between treatment arms will start to appear. T he weighted log-rank test allows a weighting for early\, middle\, and late differences through the Fleming and Harrington class of weights and is pr oven to be more efficient when the proportional hazards assumption does no t hold. We explore the impact of delayed effects in group sequential and a daptive group sequential designs and make an empirical evaluation in terms of power and type-I error rate of the of the weighted log-rank test. We a lso give some practical recommendations regarding which methodology should be used in the presence of delayed effects depending on certain character istics of the trial. \n | \n
| \n
| \n \n
John started his career at the University of Leeds before mov ing to France in the mid eighties. In the late eighties\, he worked as Ass ociate Professor of Biostatistics at the University of Washington\, Dept o f Biostatistics and the Fred Hutchinson Cancer Research Center in Seattle. Throughout the nineties until 2004\, he resided as Full Professor of math ematics at the University of California San Diego. From 2006 until 2010 he was Full Professor of Biostatistics at the University of Virginia\, since which time he was full professor at the University of Paris-Sorbonne unti l the end of 2018. In 2019 he became full professor in the Dept of Statist ical Science\, University College London. \n | \n
\n Constructing survival models
and testing effects in non-proportional hazards situations. \; \n \n | \n
| \n
| \n Dr. Satrajit Roychoudhur y is a Senior Director and a member of Statistical Research and Innovation group in Pfizer Inc. Prior to joining\, he was a member of Statistical Me thodology and consulting group in Novartis. He started his career as a res earch statistician in Schering Plough Research Institute (now Merck Co.). He has 12+ years of extensive experience in working with different phases of clinical trial. His primary expertise includes implementation of innova tive statistical methodology in clinical trial. He has co-authored several publications/book chapters in this area and provided statistical training in major conferences. His area of research includes survival analysis\, u se of model based approaches and Bayesian methods in clinical trials. Satr ajit was a recipient of a Young Statistical Scientist Award from the Inter national Indian Statistical Association in 2019. \n | \n
\n A Robust Design Appro
ach for Clinical Trials with Potential Non-proportional Hazards: A Straw M
an Proposal. | \n
|
\n
| \n \n < p>Carl-Fredrik &ldquo\;Caffe&rdquo\; Burman is Senior Statistical Science Director at AstraZeneca\, where he has been working for a quarter of a cen tury. He is part of the methodology group\, Statistical Innovation\, and w orks mainly on internal design consultations. Caffe is currently co-leadin g a project on Innovative Trial Designs for Oncology trials. He is an adju nct professor at Chalmers Univ.\n | \n\n Inference in survival trials: Weighted log-rank tests and some alternatives. \nWe can do a lot with standard t-tests\, ANCOVA and Wilcoxon. Does it have to be way more complicated if we have time-to-event data? And how can we handle trials where the relative efficacy changes over time? We will revisit some basic inference theory to generate ideas for how to test for benefit when mean survival functions may be crossing. It is demonstrated that many wei ghted logrank tests do not control the relevant type 1 error. \nWe have therefore developed the Modestly Weighted Logrank Test wit h \n1) superior power compared with the standard unweigh ted logrank test when PD-1/PD-L1 inhibitors are compared to chemotherapy\, and \n2) strong error control not only when survival fu nctions are identical but also when survival may be higher for control. \n | \n
| \n
| \n \n Martin Posch is professor of medical statistics at th e Medical University of Vienna and head of the Center for Medical Statisti cs\, Informatics and Intelligent Systems. From 2011-2012 he worked as stat istical expert at the European Medicines Agency (London\, UK) in the Human Medicines Development and Evaluation sector\, where he contributed to gui deline development and the assessment of study designs. He has a PhD in Ma thematics from the University of Vienna and was scientific assistant and a ssociate professor at the Medical University of Vienna. His research inter ests are group sequential trials\, adaptive designs and multiple testing\, focusing on applications in clinical trials and Bioinformatics. Martin Po sch serves as Associate Editor of Biometrics and Biometrical Journal and i s currently member of the executive board of the Austro-Swiss Region of th e International Biometric Society and Observer of the EMA Biostatistics Wo rking Party. \n | \n \n
Confirmative assessment of differences in the survival fu nction based on multiple characteristics. \nIf the proportional hazards assumption does not hold the standard hazard rati o estimate is not a reliable measure of the treatment effect: it depends n ot only on the actual survival distributions but also on the censoring pat tern\, the study duration\, and variations in the recruitment rates. Furth ermore\, in case of crossing hazard or survival functions estimated hazard ratios are hardly interpretable. \nHowever\, under non- proportional hazards\, differences in the survival functions can be descri bed by several parameters such as the difference between survival probabil ities at predefined time points (as 1-year and 2-year survival) the differ ence between quantiles of the survival functions (as the difference in med ians)\, or average hazard ratios computed up to a predefined time-point. W hich of these parameters is best suited to quantify the difference in the survival functions can depend on their specific shape. Therefore\, it can be desirable to specify more than one of these parameters as primary endpo int. However\, if more than one primary endpoint is considered\, a multipl icity problem arises and an inference approach controlling the family wise type I error rate as well as simultaneous coverage of confidence interval s are required for confirmatory conclusions. \nBased on the counting process representation of survival function estimators\, we s how that the considered estimators are asymptotically multivariate normal and derive an estimator of their asymptotic covariance matrix and resultin g asymptotic simultaneous confidence intervals. Furthermore\, as alternati ve method\, we derive simultaneous confidence intervals based on the pertu rbation approach for survival function estimates\, which corresponds to a parametric bootstrap. \nThe finite sample properties of the proposed methods are investigated in a simulation study. We find that coverage probabilities are close to the nominal value\, even for moderate sample sizes. \n | \n
\n Carl-Fredrik Burman\,
\n
\nCancellation and Moderation Terms
\nFo
r cancellations received up to two weeks prior to a PSI event start-date\,
the event registration fee will be refunded less 25% administrative charg
e. After this date\, no refunds will be possible. A handling fee of 20 GBP
per registration will be charged for every registration modification rece
ived two weeks prior or less\, including a delegate name change.