Discussion: View Thread

  • 1.  ITT in randomized Phase II trials

    Posted 09-08-2014 14:47
    Does anyone have any references discussing pros and cons of using intention to treat (ITT) in a first time randomized phase II comparison of a new therapy vs ann old therapy (or placebo control).  Seems to me that a per protocol (PP) analysis may sometimes be more important than ITT at this stage.  Thank you.

    -------------------------------------------

    Martin L Lesser, PhD, EMT-CC
    Director and Investigator,
       Biostatistics Unit,
       Feinstein Institute for Medical Research
    Professor, Dep't of Molecular Medicine &
         Dep't of Population Health,
       Hofstra North Shore-LIJ School of
          Medicine
    Chair, IRB Committee "B"

    -------------------------------------------


  • 2.  RE: ITT in randomized Phase II trials

    Posted 09-08-2014 15:42
    Hi Martin,

    A quick reply for now.

    Here are two references I have regarding this (hopefully helpful)
    http://www.ncbi.nlm.nih.gov/pmc/articles/PMC28218/
    http://www.sciencedirect.com/science/article/pii/S0197245600000465

    A disadvantage of PP analysis can come when looking at efficacy for example. Patient dropout or crossover may be related to treatment (i.e. bad treatment leading to dropout or crossover),
    so a PP analysis could exaggerate treatment benefits as a result. ITT might not subject to this kind of bias.


    Cheers,
    -------------------------------------------
    Jonathan L. Moscovici
    Biostatistician
    -------------------------------------------




  • 3.  RE: ITT in randomized Phase II trials

    Posted 09-08-2014 16:18
    In addition to ICH E9, Section 5.2, a couple of interesting "oldies" are
    Lewis and Machin (1993) Intention to Treat - Who should use ITT? Brit. J Cancer 63:647-650
    and
    Salsburg (1994) Intent to Treat: The reductio ad absurdum that became gospel, Pharmacoepidemiology and Drug Safety 3: 329-335.

    -------------------------------------------
    David Bristol
    Statistical Consulting Services
    -------------------------------------------




  • 4.  RE: ITT in randomized Phase II trials

    Posted 09-08-2014 16:43

    Take this as tongue in cheek, please as Sylvan intended.

    The late Sylvan Green said at a meeting:

    "There are two kinds of analysis: Intent-to-treat  and intent-to-cheat" 
    -------------------------------------------
    Jon Shuster
    University of Florida
    -------------------------------------------


    Original Message