Understanding the EMA Oncology PRO Guideline Appendix – We Can Help!

August 3, 2016 Paul O'Donohoe

EMA logo

The European Medicines Agency (EMA) recently published an appendix to their “Guideline on the Evaluation of Anticancer in Medicinal Products in Man.” The new guidance focuses on incorporating patient-reported outcome (PRO) data and health-related quality of life (HRQL) measures in oncology clinical trials. It is important to note that the appendix is intended to serve as guidance on scientific best practices, rather than a firm stance on PRO and HRQL in oncology trials.

Here are the key takeaways that I found in the appendix:

  • There are methodological obstacles that historically have reduced the impact of PRO data on regulatory decisions such as bias, missing data, quality of data, timing of assessments, single-dimensional PRO measure reporting, and lack of post-progression data.
  • PRO collection may give rise to PRO Alerts, where medically concerning levels of psychological distress or physical symptoms may require an immediate response. Where applicable, an a priori plan for the management of such alerts could be included in the protocol and communicated to trial staff.
  • Measurements should not create any undue burden to the patient. PRO experts recommend limiting estimated completion time of baseline PRO assessments to 20 minutes and 10-15 minutes for subsequent assessments.

Reading these three points immediately made me think of the benefits of electronic data capture. eCOA, in comparison to traditional paper-based questionnaires, can reduce missing data, ensure assessments are administered as per-protocol and generally improve the quality of captured data. Alerts generated from electronically captured data can, of course, be automated, allowing for timely intervention by site staff. Administration of questionnaires using a touch screen device can also reduce the level of burden experienced by sick patients compared to pen and paper. And indeed, the appendix goes on to address different modes of administration.

In line with FDA thinking on the matter, the document flags that variations in the mode of administration (i.e. mixed modes within a single study) can be a source of bias. The potential measurement bias introduced by mixing different modes could be justified if mixing modes can mitigate the more important issue of bias due to informative missing data and evidence of equivalence across different modes is available.

While speaking very positively of electronic data capture specifically, it was a shame to see the document go on to say “whilst use of electronic data recording might be of benefit in some patient groups, alternatives should be made available, e.g. for elderly patients so that differential loss of data is minimized.” CRF Health has consistently discerned positive experiences deploying eCOA in elderly populations and, as long as sufficient training is provided, the elderly can often be one of our most compliant populations.

As you can see, there are some challenges and risks associated with collecting PRO data in oncology clinical trials, but they can be readily addressed with a variety of solutions, including electronic administration of clinical outcome assessments (eCOA).

Here are some additional resources to check out:

What were your key takeaways from the new guidance? Let us know!

About the Author

Paul O'Donohoe

Director of Health Outcomes at CRF Health

Previous Guide
Reducing the Burden of Data Collection in Oncology Trials
Reducing the Burden of Data Collection in Oncology Trials

Next Presentation
Putting the Oncology Patient First
Putting the Oncology Patient First

With almost half of oncology studies failing due to a lack of patient retention, there is a critical need t...