Background: Eliminating hepatitis C is hampered by the costs of direct-acting antiviral treatment and the need to treat hard-to-reach populations. Access could be widened by shortening or simplifying treatment, but limited research means it is unclear which approaches could achieve sufficiently high cure rates to be acceptable. We present the statistical aspects of a multi-arm trial designed to test multiple strategies simultaneously and a monitoring mechanism to detect and stop individual randomly assigned groups with unacceptably low cure rates quickly.
Methods: The VIETNARMS trial will factorially randomly assign patients to two drug regimens, three treatment-shortening strategies or control, and adjunctive ribavirin or no adjunctive ribavirin with shortening strategies (14 randomly assigned groups). We will use Bayesian monitoring at interim analyses to detect and stop recruitment into unsuccessful strategies, defined by more than 0.95 posterior probability that the true cure rate is less than 90% for the individual randomly assigned group (non-comparative). Final comparisons will be non-inferiority for regimens (margin 5%) and strategies (margin 10%) and superiority for adjunctive ribavirin. Here, we tested the operating characteristics of the stopping guideline for individual randomly assigned groups, planned interim analysis timings and explored power at the final analysis.
Results: A beta (4.5, 0.5) prior for the true cure rate produces less than 0.05 probability of incorrectly stopping an individual randomly assigned group with a true cure rate of more than 90%. Groups with very low cure rates (<60%) are very likely (>0.9 probability) to stop after about 25% of patients are recruited. Groups with moderately low cure rates (80%) are likely to stop (0.7 probability) before overall recruitment finishes. Interim analyses 7, 10, 13 and 18 months after recruitment commences provide good probabilities of stopping inferior individual randomly assigned groups. For an overall true cure rate of 95%, power is more than 90% to confirm non-inferiority in the regimen and strategy comparisons, regardless of the control cure rate, and to detect a 5% absolute difference in the ribavirin comparison.
Conclusions: The operating characteristics of the stopping guideline are appropriate, and interim analyses can be timed to detect individual randomly assigned groups that are highly likely to have suboptimal performance at various stages. Therefore, our design is suitable for evaluating treatment-shortening or -simplifying strategies.
Trial registration: ISRCTN registry: ISRCTN61522291. Registered on 4 October 2019.
Keywords: Adaptive design; Bayesian methods; Clinical trial; Hepatitis C; Interim analyses; Multi-arm; Trial design.