Case Study

 

Adaptive Randomization

When looking for a powerful and sophisticated IRT solution to support complex randomization, a pharmaceutical sponsor chose YPrime’s IRT+. The sponsor needed a custom solution to work across multiple protocols within the same program, involving approximately 3,000 patients with conjunctivitis. A major requirement specified the use of adaptive randomization instead of a traditional, pre-determined randomization list. Adaptive randomization would ensure that subjects would be assigned among the different treatment groups in a balanced manner, according to their treatment group ratios.

As conjunctivitis is extremely contagious, it was important to the sponsor that subjects from the same household were randomized to the same treatment group. This requirement would help avoid risk of cross-treatment (i.e. one subject using another subject’s study drug especially if those two study drugs were different treatment types). If the sponsor left randomization to chance in a traditional, pre-determined randomization list, the randomization ratio for each treatment group would likely be imbalanced, and therefore detrimental to data integrity.

In response, YPrime developed a system feature that randomized participants within the same household into the same treatment groups without creating a ratio imbalance.

During the design phase of the study, 500 simulations were run for three scenarios for each of the three studies at two different timepoints to ensure the adaptive randomization was performing as expected.

This project formed the basis of a longer-term partnership and the unique feature can now be applied to similar studies within the same clinical program.