BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Drupal iCal API//EN
X-WR-CALNAME:Events items teaser
X-WR-TIMEZONE:America/Toronto
BEGIN:VTIMEZONE
TZID:America/Toronto
X-LIC-LOCATION:America/Toronto
BEGIN:DAYLIGHT
TZNAME:EDT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
DTSTART:20190310T070000
END:DAYLIGHT
BEGIN:STANDARD
TZNAME:EST
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
DTSTART:20191103T060000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
UID:69f416b022629
DTSTART;TZID=America/Toronto:20200204T100000
SEQUENCE:0
TRANSP:TRANSPARENT
DTEND;TZID=America/Toronto:20200204T100000
URL:https://uwaterloo.ca/statistics-and-actuarial-science/events/department
 -seminar-haolun-shi-simon-fraser-university
LOCATION:M3 - Mathematics 3 200 University Avenue West Room 3127 Waterloo O
 N N2L 3G1 Canada
SUMMARY:Department seminar by Haolun Shi\, Simon Fraser University
CLASS:PUBLIC
DESCRIPTION:BAYESIAN UTILITY-BASED TOXICITY PROBABILITY INTERVAL DESIGN FOR
  DOSE\nFINDING IN PHASE I/II TRIALS\n\nMolecularly targeted agents and imm
 unotherapy have revolutionized\nmodern cancer treatment. Unlike chemothera
 py\, the maximum tolerated\ndose of the targeted therapies may not pose si
 gnificant clinical\nbenefit over the lower doses. By simultaneously consid
 ering both\nbinary toxicity and efficacy endpoints\, phase I/II trials can
  identify\na better dose for subsequent phase II trials than traditional p
 hase I\ntrials in terms of efficacy-toxicity tradeoff.  Existing phase I/
 II\ndose-finding methods are model-based or need to pre-specify many\ndesi
 gn parameters\, which makes them difficult to implement in\npractice. To s
 trengthen and simplify the current practice of phase\nI/II trials\, we pro
 pose a utility-based toxicity probability interval\n(uTPI) design for find
 ing the optimal biological dose (OBD) where\nbinary toxicity and efficacy 
 endpoints are observed. The uTPI design\nis model-assisted in nature\, sim
 ply modeling the utility outcomes\nobserved at the current dose level base
 d on a quasibinomial\nlikelihood. Toxicity probability intervals are used 
 to screen out\noverly toxic dose levels\, and then the dose escalation/de-
 escalation\ndecisions are made adaptively by comparing the posterior utili
 ty\ndistributions of the adjacent levels of the current dose. The uTPI\nde
 sign is flexible in accommodating various utility functions while\nonly ne
 eds minimum design parameters. A prominent feature of the uTPI\ndesign is 
 that it has a simple decision structure such that a concise\ndose-assignme
 nt decision table can be calculated before the start of\ntrial and be used
  throughout the trial\, which greatly simplifies\npractical implementation
  of the design. Extensive simulation studies\ndemonstrate that the propose
 d uTPI design yields desirable as well as\nrobust performance under variou
 s scenarios. This talk is based on the\njoint work with Ruitao Lin and Yin
 g Yuan at MD Anderson Cancer Center.\n 
DTSTAMP:20260501T025752Z
END:VEVENT
END:VCALENDAR