Adaptive platform trials are needed for more informed policy decisions on antiviral therapies

A research commentary by a CITF-funded researcher published in the Canadian Medical Association Journal discusses how the study populations enrolled in research on oral antivirals (e.g., nirmatrelvir-ritonavir, molnupiravir) are not representative of current COVID-19 patients. The authors argue that this key shortcoming means that these therapies, aimed at reducing death and hospitalizations, are less effective in real-world settings. They suggest adaptive platform trials could serve to overcome this shortcoming. The research was led by Dr. Corinne Hohl (University of British Columbia).

The study populations in the large trials of oral antiviral therapy were mostly unvaccinated individuals infected with the Delta variant who had no previous SARS-CoV-2 infection. At the time, 83% of the eligible Canadian population had received 2 doses of SARS-CoV-2 vaccine, conferring a high degree of protection against the same outcomes these therapies were found to prevent. In fact, when antiviral therapy trials have included vaccinated populations, they have shown vaccination itself to be as effective, if not more, than antivirals like nirmatrelvir-ritonavir.

The authors believe that to be effective, clinical and policy decisions about antiviral therapies should be informed by research applicable to current patient characteristics. However, they point to the lack of randomized controlled trials with high-risk and under-represented subgroups (e.g., immunocompromised, pregnant, or older people).

To guide policy decisions, the authors recommend national and international use of adaptive platform trials, because:

• They can iteratively integrate novel therapies and large-scale observational studies using linked administrative healthcare data to estimate outcome risks in important clinical subgroups. These represent the best options for rapidly evaluating the effectiveness of novel antiviral drugs and other COVID-19 therapeutics as both the virus and population characteristics evolve.
• Adaptive platform trials can be conducted successfully in Canada if there is dedicated funding support, along with policy changes to facilitate data sharing by administrative data custodians and integration of trial enrolment at the point of care.
• Real-world, post-marketing observational studies that are linked to administrative health care data can generate adequately powered estimates of effectiveness in subgroups of patients.