Correcting for non‐compliance in randomized non‐inferiority trials with active and placebo control using structural models

Abstract

The three-arm clinical trial design, which includes a test treatment, an active reference, and placebo control, is the gold standard for the assessment of non-inferiority. In the presence of non-compliance, one common concern is that an intent-to-treat (ITT) analysis (which is the standard approach to non-inferiority trials), tends to increase the chances of erroneously concluding non-inferiority, suggesting that the per-protocol (PP) analysis may be preferable for non-inferiority trials despite its inherent bias. The objective of this paper was to develop statistical methodology for dealing with non-compliance in three-arm non-inferiority trials for censored, time-to-event data. Changes in treatment were here considered the only form of non-compliance. An approach using a three-arm rank preserving structural failure time model and G-estimation analysis is here presented. Using simulations, the impact of non-compliance on non-inferiority trials was investigated in detail using ITT, PP analyses, and the present proposed method. Results indicate that the proposed method shows good characteristics, and that neither ITT nor PP analyses can always guarantee the validity of the non-inferiority conclusion. A Statistical Analysis System program for the implementation of the proposed test procedure is available from the authors upon request.

An overview of statistical methods for handling nonadherence to intervention protocol in randomized control trials: a methodological review

“…60 Another version of SMM applied to accelerated failure time (AFT) model (time to event survival data with time as outcome) is Rank Preservative Structural Failure Time model (RPSFTM). 61,62 They are called rank preserving because they use a class of rank estimators for subjects' failure. 62 In practice, G-estimators have not been widely adopted due to level of complexities involved in implementation.…”

“…61,62 They are called rank preserving because they use a class of rank estimators for subjects' failure. 62 In practice, G-estimators have not been widely adopted due to level of complexities involved in implementation. 63 For handling non-adherence in continuous time survival data include Cox-reg (1,2) , Complier proportional hazard effect of treatment (C-PROPHET) model and Causal accelerated life model (CALM).…”

What are the statistical implications of treatment non‐compliance in cluster randomized trials: A simulation study

“…In the most recent version of the CONSORT statement, the specific request for ITT has been dropped in favour of a clear description of exactly who was included in each analysis . Wu et al conclude that neither ITT nor PP can always guarantee the validity of the conclusion in non‐inferiority trials …”

“…12 Wu et al conclude that neither ITT nor PP can always guarantee the validity of the conclusion in non-inferiority trials. 13 Given the variety of approaches to deal with non-compliance and recommendations of which approach to use, it is important to understand how each of them performs. Only then, a motivated choice can be made for a trial at hand.…”

Sequential parallel comparison design for “gold standard” noninferiority trials with a prespecified margin

“…The second is the so‐called “fixed‐margin approach,” which defines Δ as a prespecified difference between E and R (Koch & Röhmel, ). There have been numerous publications on the design and analysis for the fraction approach (Dette, Trampisch, & Hothorn, ; Ghosh, Nathoo, Gönen, & Tiwari, ; Hasler, Vonk, & Hothorn, ; Hauschke & Pigeot, ; Hung, Wang, & O'Neill, ; Kieser & Friede, ; Li & Gao, ; Mielke, Munk, & Schacht, ; Mütze, Konietschke, Munk, & Friede, ; Mütze, Munk, & Friede, ; Pigeot et al., ; Wu, Li, Hou, Li, & Zhou, ; Wu, Zhao, Hou, Li, & Zhou, ; Zhou & Kundu, ) and for the fixed‐margin approach (Ghosh, Ghosh, & Tiwari, ; Hida & Tango, ; Hida & Tango, ; Hida & Tango, ; Huang, Wen, & Cheung, ; Koch & Röhmel, ; Kwong, Cheung, & Hayter, ; Kwong, Cheung, Hayter, & Wen, ; Lu, Jin, & Zeng, ; Mütze & Friede, ; Röhmel & Pigeot, ; Schlömer & Brannath, ; Stucke & Kieser, ; Stucke & Kieser, ; Zhong, Wen, Kwong, & Cheung, ). The other issue is how to obtain the proof of assay sensitivity in gold standard noninferiority trials.…”