Master protocol: umbrella-trial-based platform trial versus basket-trial-based platform trial

Biotech company Aerami Therapeutics announced that they are going to design a platform trial to test their pulmonary hypertension drug simultaneously in multiple indications (in this case, multiple WHO groups of pulmonary hypertension).

uniPHied: A Novel Platform Study Design for a Phase 2 Clinical Trial to Assess the Safety and Efficacy of AER-901 in Pulmonary Hypertension Associated With Interstitial Lung Disease and Pulmonary Arterial Hypertension

They stated the following "Phase 2 studies in pulmonary hypertension have been confined to one etiology, with large participant numbers relative to rare disease demographics. This platform study design allows for assessment of AER-901 across multiple forms of pulmonary hypertension within one efficient trial. The uniPHied study aims to assess AER-901 in pulmonary hypertension associated with interstitial lung disease (PH-ILD) and pulmonary arterial hypertension (PAH), with the option to expand to other forms of pulmonary hypertension via a protocol amendment".

Pulmonary Hypertension (PH) comes in multiple forms (phenotypes or genotypes). According to the WHO Pulmonary Hypertension classification, there are five different groups (see the table below). Each WHO PH class is considered a separate indication. Drug development programs in PH will include clinical trials for each WHO PH class. 

It is natural to think of a master protocol that can combine the clinical trials in different WHO PH classes into a single clinical trial (a platform trial). 

However, there are two issues in the uniPHied study design. 

This is not a typical platform trial design. the platform trial is usually based on an umbrella trial where multiple drugs are evaluated for the same indication. The platform trial has not been used for basket trials where the same drug is evaluated for multiple indications (here in multiple WHO PH classes). Here are the definitions for master protocol, umbrella trial, platform trial, and basket trial. 

FDA's guidance for industry "Master Protocols for Drug and Biological Product Development"

Master protocol: a protocol designed with multiple substudies, which may have different 22 objectives and involve coordinated efforts to evaluate one or more medical products in 23 one or more diseases or conditions within the overall study structure.   

Substudy: the information and design features (e.g., objectives, design, methodology, 26 statistical considerations) related to evaluation of a single medical product in a single 27 disease, condition, or disease subtype in the master protocol.  

     

The control group can not be shared across different sub-studies in different WHO PH classes. 

The uniPHied study is a basket-trial-based platform trial. Each indication (each WHO PH classes) needs to have its own control group. The data from control groups in different WHO PH classes can not be combined or pooled.

The platform trial in the FDA's guidance refers to the umbrella-trial-based platform trial. FDA guidance discussed the shared control group extensively. The control group in the umbrella-trial-based platform trial needs to be separated into a concurrent control and a nonconcurrent control. For a specific drug in the platform trial, FDA recommended the following: 

The control group used for the primary comparison of any given drug in a master protocol should generally include only concurrently randomized subjects (i.e., a concurrent control) and should not include nonconcurrently randomized subjects.