Clinical Trials Arena spoke to Anthony T. Everhart, MD, FACP; Clinical Vice President at leading evidence generation company, Signant Health, about recent regulatory draft guidance on the use of patient-reported outcome measures (PROMs) in oncology trials and recommendations for trial design and conduct.

Q: Sponsors are increasingly using PRO data to support clinical trial endpoints. What are some of the key challenges they face when collecting this data?

Anthony: While the traditional way of collecting clinical trial data has been through the use of paper-based methods, using electronic data capture systems allows the data collected to be attributable, legible, and contemporaneous.

When using an electronic clinical outcome assessments (eCOA), sponsors can identify when the data was collected as it is dated and time stamped. Additionally, alerts and edit checks can keep instruments from being completed incorrectly and prevent blank fields resulting in high-quality data1. eCOA can be even more helpful when monitoring the quality of data collected remotely. The addition of alerts and reminders in an eCOA solutions can improve patient adherence with assessment completion, while the capture of date and time stamps allows investigators to determine exactly when assessments were completed. This reduces the “parking lot effect” where patients complete their entire paper diary in the parking lot just before their visit, relying on recall over longer periods rather than providing contemporaneous data according to the design of the assessment2.  

Q: How can eCOA help collect patient data in an oncology research study?

AE:  Many oncology studies, particularly early phase studies, complete COA on paper. However, items can be left blank, they can be scratched out, or even illegible. Electronic data capture on-site improves the quality of data and the time of data collection. With electronic capture of data through diaries or standardised assessments remotely, we open up the possibility to collect data more frequently from patients in-between visits. This allows us to add even more of the voice of the patient into the study by collecting data at home between treatment cycles, compared to the traditional on-site, start-of-cycle method of collection that we’ve been using for many years.

Q: The FDA recently published draft guidance on the selection and implementation of PROs in oncology clinical trials. Can you highlight the guidance around COA measurement strategies?

AE: Over the past decade or more, we have seen very little PRO data make their way into the final labelling of the drug. There are several reasons for this. The first relates to the heterogeneity of the assessments in previous trial designs, and the second to the instruments themselves having a fair amount of domain overlap and lacking specificity. The third factor relates to the large amount of missing data seen in previous submissions. For those reasons, there has been hesitancy to include PRO information and data in the reviews, analysis, and labelling of new drugs.

In response to this, the FDA has released draft guidance outlining requirements for core patient-reported outcomes in cancer clinical trials. One of the recommendations is for the standardisation of collecting disease-related symptoms. One way to do this would be through the use of an instrument such as the PRO-CTCAE. It also recommends a single-item assessment for overall side effect impact as well as more domain specificity within standardized instruments, specifically focused on both physical and role function. Further to this, the FDA recommends more frequent collection of PROs and potentially collecting this data at a more rapid cadence during the early phases of treatment and tapering off to slower cadence later in the maintenance phase or follow-up phase.  Doing this would require a shift from site-based assessments to remote assessments.

The guidance also recommends the collection of the reasons for missing data. In the case where a patient is unwell, or even hospitalised, and cannot answer their PRO assessments, the absence of those patient-reported outcomes and reason for the missing data is still important information. Rather than reviewing an incomplete data set, the FDA can use the reason for missing data to inform their analysis and decision making.

Q: What role does technology play in trial optimisation?

AE: Trial optimisation begins with protocol design. For around 30 years now, we’ve been designing protocols the same way. During the COVID-19 pandemic, we were forced to rapidly decentralise studies which had been designed to be completed on-site. However, continuing to design site-based protocols and then “decentralise” them is inefficient.  What we should be focusing on is trial optimization.  

While this idea seems simple, it can be hard to achieve. An optimised study collects the right data at the right time with the right solution in the right setting. Trying to capture data that’s better collected on-site at home, or vice versa, results in a study which may be less than optimised. Using technology such as electronic capture for patient reported outcomes, sensors and wearables to collect other objective measures of patient performance, and telemedicine can decrease patient burden can support trial optimisation.

Q: What does the future of oncology trials look like?

AE: The shift that we’re going to see in oncology trials revolves around the movement from on-site collection of information to remote collection of data. We will see more frequent remote collection of quality-of-life assessments and symptom reporting. We are likely to see increased use of sensors and wearables for measuring a patient’s activity levels or sleep quality as indicators of disease progression or tolerability of treatment.

I also expect to see increased use of remote e-consent as we try to expand our network of patients who could participate in research, getting potential research protocols in front of a broader audience so they can become informed about a study. We may even see more of a spoke-and-hub model of research where we have central investigators running a study in concert with satellite facilities and sub-investigators in order to reach a broader population using electronic tools to facilitate.


1 Clinical Viewpoints video series episode 3:

2 Stone AA, Shiffman S, Schwartz JE t al. Patient non-compliance with paper diaries. British Medical Journal 2002; 324: 1193-1194.

Discover more about how to optimise your COA measurement strategy by downloading the free whitepaper below.