Combining Factorial and MAMS designs

06 Jul, 2022

The methodology team at the MRC Clinical Trials Unit at UCL is a key partner of Penta and has made a huge contribution to our studies on HIV. The team recently published a paper in Sage Journals describing how the advantages of two designs can be practically combined:

  • factorial designs, which do two or more randomisations in the same participants, and
  • MAMS platform designs, which may “drop” or add arms during the course of a trial.

The combined design allows testing of multiple primary research hypotheses, which should lead to faster decisions about experimental treatments.

The main features of the combined design are

  • multiple research treatments can be compared against the standard of care within the framework of factorial designs;
  • over the course of the trial, further research arms can be added to different randomisations in the factorial component;
  • new randomisations can be introduced;
  • recruitment to some of the research treatments can be stopped early, possibly leading to some randomisations being entirely dropped.

The paper discusses analysis and conduct considerations in this new setting, including:

  • how to time the interim and final analyses;
  • the importance of broad eligibility criteria;
  • how to implement randomisations sequentially;
  • how to perform a valid analysis.

It also shows some ways to display the evolution of a factorial-MAMS trial over time (see Figure below).

The paper is authored by Ian R White, Babak Choodari-Oskooei, Matthew R Sydes, Brennan C Kahan, Leanne McCabe, Anna Turkova, Hanif Esmail, Diana M Gibb, Deborah Ford. Combining factorial and multi-arm multi-stage platform designs to evaluate multiple interventions efficiently. Clinical Trials. May 2022. doi:10.1177/17407745221093577


Photo by megaflopp from Getty Images