Collection of Patient-Reported Outcomes Feasible in Cancer Clinical Trials

May 10, 2017, by NCI Staff

When patients with cancer participate in clinical trials of new therapies, it's usually the clinicians who assess and record the side effects that the patients experience.

But a new study has found that patients, even those who are undergoing difficult treatments, are willing to devote time to completing thorough assessments of the side effects they encounter.

The study, in fact, found that most gaps in these patient-reported outcomes data—which were captured during clinic visits using tablet computers—weren't caused by patients feeling too ill to provide reports. Instead, they were due largely to technical errors like lost Internet connectivity.

"Patients who are participating in clinical trials are enthusiastic about providing information about how they're feeling, and they value the opportunity to have that information become part of the study," said the study's lead investigator, Ethan Basch, M.D., of the Lineberger Comprehensive Cancer Center at the University of North Carolina.

With technological advances, such as newer tablet computers and smart phones, as well as broader access to the Internet, he added, patient-reported outcomes (PROs) should be able to provide a much fuller picture of how new drugs affect patients' quality of life.

The results were published in the June 1, 2017 International Journal of Radiation Biology * Oncology * Physics.

Gaps in Understanding

Side effects experienced by participants in NCI-sponsored clinical trials are logged into the Common Terminology Criteria for Adverse Events (CTCAE) system. This tool allows reports on everything from blood test values to toothaches. It also includes many symptoms—such as nausea, anxiety, and nerve pain—that are by their nature personal and difficult for an observer to measure.

As part of their standard patient assessments, staff involved in trials use the CTCAE system to record any side effects patients are experiencing. But researchers have long wondered if creating a way for patients to document their own experiences during treatment would help to provide more complete and accurate information about symptomatic side effects that may be associated with new drugs or treatment regimens.

"Patient reporting in various forms could improve the quality of side-effect data collection in clinical trials," said Dr. Basch. This is important in drug development, he explained, because if researchers miss or underestimate the severity of side effects caused by a drug they can inadvertently pick the wrong dose or the wrong treatment schedule.

"Or we can put something out on the market without adequately warning people about what to expect, and then people can get walloped with the side effects and don't want to take the drug," he continued. This can lead to patients stopping potentially effective drugs, particularly if they're oral drugs taken at home without direct supervision by a health care provider, he added.

Cancer care providers saw this happen with aromatase inhibitors, explained Dr. Basch. Clinical trials of these breast cancer drugs underestimated the prevalence of joint and muscle pain that patients experienced.

As a result, up to 40% of women "stop taking the drugs very soon after starting, and another 20% will stop taking them within the first year or two," he said, rather than taking them for the recommended 5- or 10-year course.

Capturing the Patient Experience

Researchers at NCI and at cancer centers across the United States set out to build a version of the CTCAE system to be used for self-reporting of symptomatic side effects by patients, called the PRO-CTCAE.

The team combed through the CTCAE to identify those adverse events that are amenable to self-reporting. They identified 78 symptomatic side effects that are associated with common treatments, including chemotherapy, radiation therapy, biological therapy, and immunotherapy, explained Sandra Mitchell, Ph.D., program director in NCI's Outcomes Research Branch and current scientific lead for PRO-CTCAE development. A library of questions phrased using plain, patient-friendly language was developed to capture these side effects directly from patients, Dr. Mitchell continued.

"We then validated these questions using a national sample of patients who were undergoing cancer treatment, to ensure that the questions were easily understood and measured the side effects of interest in a consistent manner," she said.

Previous studies had shown that the PRO-CTCAE system could reliably capture what patients experienced during treatment. To test whether it could be used in the clinical trial setting, the research team embedded PRO-CTCAE in a clinical trial that enrolled patients undergoing chemotherapy and radiation therapy for lung cancer.

All 152 patients in the trial, from 80 participating hospitals, were asked to complete a subset of the PRO-CTCAE that contained 53 separate items at several scheduled visits: before treatment, weekly during the 4 weeks of radiation therapy, and once after treatment.

Overall, patient-reported outcomes were successfully collected at 86% of scheduled visits. The most common reasons that the data could not be collected were missed clinic appointments, simple oversights (such as forgetting to provide tablet computers to participants), and technical problems (such as computer malfunctions or Internet connectivity issues). Only 13% of missed reports occurred because patients felt too sick to fill out the forms.

"This is feasible to do—even sick patients in various clinical settings are going to be able to complete [PRO-CTCAE reports], and I think that's extremely encouraging, even though there are logistical issues to consider," said Dr. Mitchell.

Based on lessons learned during this study and a previous study of the implementation of patient-reported outcomes in a group of NCI-sponsored clinical trials, Dr. Basch and his colleagues are conducting another study of the PRO-CTCAE in a large rectal cancer trial.

In that study, "we moved reporting out of the clinic entirely," he said. Patients reported their adverse events from home every week, either online or using an automated phone system. If a participant missed a report, they received a call from a data manager to check in on them.

This human contact increased compliance with self-reporting, he added. The full results from that study are being prepared for publication.

In addition to its use in NCI-sponsored clinical trials, Dr. Basch sees potential for using PRO-CTCAE in studies that are comparing the tolerability of different drugs, as well as in routine cancer care.

"The PRO-CTCAE was designed and developed to do all of these things, and I think it's up to us as a community to now make use of this tool to try to improve our patients' experience," he concluded.

Ongoing research is focused on the optimal ways to analyze and interpret PRO-CTCAE scores in the context of cancer clinical trials, and on the development and testing of versions translated into other languages, said Dr. Mitchell.