RAPS is closely monitoring developments in the Coronavirus (COVID-19) outbreak. See our public safety page for the latest updates.

Regulatory Focus™ > News Articles > 2020 > 9 > How to assess symptoms in COVID-19 trials: new FDA guidance

How to assess symptoms in COVID-19 trials: new FDA guidance

Posted 16 September 2020 | By Kari Oakes 

How to assess symptoms in COVID-19 trials: new FDA guidance

Investigators should use a patient-reported outcomes approach, with an emphasis on electronic data collection, in their research on therapies to treat or prevent COVID-19, according to a new guidance from the US Food and Drug Administration (FDA).

The guidance, which is effective immediately during the public health emergency of the coronavirus pandemic, outlines how to capture disease-related symptoms in clinical trials focusing on COVID-19 prevention or treatment for outpatient adults and adolescents.   

Using patient-reported outcome (PRO) instruments “is advised when measuring signs and symptoms best known by the patient or best measured from the patient perspective,” the agency says in the guidance. Such signs and symptoms might include shortness of breath, pain and fatigue, for example. Sponsors should consult FDA about which PRO instruments to use. Whatever tool an investigator chooses should be evaluated for basic comprehensibility and usability.

Nuts and bolts of PRO assessments
PRO assessments should be obtained at least daily, and at the same time each day, according to the guidance. The best option would be to use electronic data collection systems that can send reminders and time-stamp data collection, but if a paper-based data collection diary is used, patients should still receive calls, texts or emails to remind them to fill in their data daily.

All participants should be asked about a standard set of COVID-19‒related symptoms, regardless of what symptoms they were experiencing at baseline, because new symptoms may emerge after the baseline assessment, the guidance says. It emphasizes that symptoms of the novel coronavirus can be heterogeneous and unpredictable in outpatients.

The guidance provides an example of a structured instrument that assesses 14 key COVID-19‒related symptoms, with suggested weighting and scales. FDA recommends avoiding excessively long PRO instruments because the subject burden is increased, as is the likelihood of missing data.

The guidance also includes some general considerations for PRO instrument design and implementation. These include using verbal descriptors such as “mild,” “moderate” and “severe” in response scales to questions about specific symptoms. The rationale, said the guidance, is that “the absence of verbal descriptors may create difficulty in interpretation in this context of use.”

Additional assessments that can be considered for inclusion in clinical trials of drugs to treat or prevent COVID-19 include details of analgesic and antipyretic use; body temperature, including the provision of thermometers; oxygen saturation data, with pulse oximeters provided to participants; and patient-reported global impression items to assess return to usual health, activities, and overall symptom burden.

In terms of endpoints, FDA refers sponsors to a May 2020 guidance, titled COVID-19: Developing Drugs and Biological Products for Treatment or Prevention, that suggests time to sustained clinical recovery, assessed over “an appropriate duration,” would be a suitable endpoint.
Investigators can consider that a participant has sustained clinical recovery when COVID-19‒related symptoms identified as key do not score above a prespecified threshold over a “clinically meaningful time.”

Sponsors, says the guidance, should have entry criteria for their trials that define a minimal baseline severity score for symptoms so that they can detect and accurately evaluate clinical benefit of a therapy under investigation. Similarly, trial screening criteria should exclude participants who need hospitalization.

The guidance cautions against some tactics in trial design, including defining endpoints in terms of an aggregate score obtained by adding items within the set of common symptoms. The agency also does not recommend calculating an area under the curve for COVID-19‒related symptoms. “Given the heterogeneity of COVID-19‒related symptoms, any single trial subject may only experience a small subset of the common COVID-19‒related symptoms described in this guidance,” notes the agency.

Data management
Sponsors should prospectively plan how to handle missing data in trials for COVID-19 treatments or prophylactic interventions, including how hospitalization and death will be handled during data analysis.

However, the goal should be to minimize missing data through a well-designed trial that includes plenty of opportunities for participants to receive reminders and real-time compliance monitoring and include the option to capture responses verbally, says the guidance. Sponsors can also reinforce the importance of adherence to trial protocols during the informed consent process.

“The aim should be to ascertain vital status for all COVID-19 trial subjects even after a subject decides to discontinue treatment or discontinue participation in the trial, including follow-up for key outcomes, while adhering to informed consent requirements,” said FDA.


Tags: COVID-19, FDA

Regulatory Focus newsletters

All the biggest regulatory news and happenings.