A questionnaire given to cancer patients involved in clinical trials is able to accurately identify a wide range of adverse events after treatment, according to researchers at the University of North Carolina's Lineberger Cancer Outcomes Research Center, who were asked to assess the validity and reliability after developing the tool for the National Cancer Institute.
The results of the study, now online ahead of print in JAMA Oncology (doi:10.1001/jamaoncol.2015.2639), showed that the tool more accurately captures adverse events (AEs) in light of studies that have found discrepancies between side effects reported by physicians and those reported directly by patients. For example, such gaps in perception and reporting differed by as much as 74 percent in a series of European trials.
The lead author, Ethan Basch, MD, Director of the Cancer Outcomes Research Program and Associate Professor of Medicine and Public Health at the UNC School of Medicine, explained that the system supplements the CTCAE (Common Terminology Criteria for Adverse Events), which clinical trial researchers have used to document adverse events. While the CTCAE is meant to capture all AEs experienced by patients, previous studies have shown that it does not fully capture the wide range of side effects experienced by patients, nor many details about the frequency and severity of the side effects as well as the impact on patients' daily lives.
78 Potential Side Effects
For the new study (first author is Amylou C. Dueck, PhD, Head of the Section of Biostatistics at Mayo Clinic in Scottsdale, Arizona) of the patient-reported outcomes version of the tool (PRO-CTCAE), the researchers tested all 124 items included to assess 78 potential side effects from chemotherapy and/or radiotherapy, with the findings reported based on answers from patients at nine cancer centers or individual oncology offices.
Data were collected on 940 out of 975 subjects at their first visit, and 852 of 940 participants at their second visit, with visits about six weeks apart. The patients' median age was 59; 57 percent were women; 32 percent had a high school education or less; and 17 percent had an ECOG performance status of 2 to 4.
Nearly all the patients (99.8%) reported at least one symptom they rated as being frequent, severe, or impacting their daily living “quite a bit.” Patients were each experiencing a median of 23 symptoms (range of 0 to 91), with 904 of 940 (96.2%) reporting five or more at the first visit.
The most often reported physical AEs were vomiting/nausea, loss of appetite, fatigue, dry mouth, loose stool, mouth or throat sores, swelling of the arms or legs, general pain, headaches, and shortness of breath. The most common psychological AEs were anxiety, insomnia, and sadness or depressive symptoms.
‘Positive Response from NCI’
Basch said in an interview that he considers this is to be a landmark study because patient-reported information has generally not been used for measuring side effects in cancer research, and the findings show that the tool is effective for use in clinical trials.
“The response from the NCI has been very positive, and it is reviewing the data, while the U.S. Food and Drug Administration has said that it should be integrated into all oncology trials,” Basch said.
Spanish & Other Language Versions
A Spanish version of the questionnaire is already available, and efforts are underway to develop additional translations, Basch added. Moreover, some clinical trial researchers in the U.S. have already received permission from the NCI to use the system. A similar system is already being used in Japanese trials, he noted.
“Although we only reported the results of several follow-up visits, I believe it should be used at every subsequent office or clinic visit.”
He explained that during office visits, questions about side effects are typically limited due to time constraints and other factors, so the frequency, severity, and range of side effects are often not fully reported, with the result that AEs are underestimated. “Patients often feel worse, but their physicians are not aware of the extent of their AEs and how much they impact not only quality of life but also compliance to oral medications prescribed.”
Due in part to this risk, it is important that greater effort is made to improve patient and physician communication to make sure all AEs are reported as well as their impact on daily activities, Basch said.
‘Unprecedented in a Validation Study’
Also commenting in an interview, Dueck said: “A major achievement of this study is that we tested over 100 distinct questions about adverse events, which is unprecedented in a validation study.
“We ultimately were able to validate 119 out of the 124 items in the PRO-CTCAE in over 900 racially diverse patients across the U.S., including many patients with impaired physical functioning, and with myriad cancer types. This wide representation of patients better reflects the spectrum of people enrolled in cancer clinical trials.”
Accompanying Commentary by Benjamin Movsas
In an accompanying commentary (doi: 10.1001/jamaoncol.2015.2689), Benjamin Movsas, MD, Chair of Radiation Oncology at Henry Ford Hospital and a leader in several nationwide, multicenter clinical trials, noted that for more than 30 years oncology trials have relied on the NCI's CTCAE despite reports that there is often a “disconnect” between what researchers report and the actual experience of trial subjects.
“This begs the question: When it comes to reporting symptomatic adverse events, should the perspective of the patient or the clinician be primarily considered?
“Some would argue that the clinician is most qualified to report AEs and/or say that patient-reported outcomes are not scientifically rigorous in clinical trials because they are based on subjective reporting. At a fundamental level, though, how can anyone know the patient's subjective experience better than the patient?”
The researchers who developed and evaluated the new system “took on a daunting task,” Movsas continued. “However, while the study is an important first step, more work is needed. For example, less than four percent of the patients in this study underwent cancer surgery, so this group requires further study.”
In addition, because the reliability data were limited to a subset of AEs, more studies will be needed to analyze the reliability of the test, he said.
“Practical issues will also need to be addressed regarding how PRO-CTCAE may affect administrative time, cost, and patient burden over time. Beyond logistic issues, the ultimate success of the PRO-CTCAE will depend on imparting its importance and relevance to patients, clinicians, and other stakeholders.”
Asked for his perspective for this article, Tomasz M. Beer, MD, the Grover C. Bagby Endowed Chair for Prostate Cancer Research and Director of the Prostate Cancer Research Program at Oregon Health & Science University, was enthusiastic about the findings: “I am very excited about this because to be complete, our clinical trial data should include both physician-reported side effects of treatments and those by patients. Differences between physician and patient reports of AE data can happen for many reasons, but we need both sides.”
Even so, he said he can't help but be concerned about how well the system will work in real life: “This is a long questionnaire, and patients tend to get tired of answering questions about symptoms that they have already reported. We have seen that some patients start skipping their responses, even after two or three weeks. Getting complete and accurate data at every office visit will be a challenge, and I think we still have to see how this plays out.”
Another challenge, he said, is that patients are not always asked about all their adverse events in the same way, so implementation will require patient education about the importance of the system. Patients do not always report all of the problems they experience from treatment, and if they do not report all AEs, including anxiety and depression, along with information about how any symptoms are impacting their lives, clinical trials will miss these.
“We will need to ensure that research oncologists emphasize to patients the importance of answering all of the questions. So with the caveats about implementation and adherence challenges, the approach of involving patient in AE reporting holds great promise. The new system will provide a better understanding of AEs in these patients and provide greater consistency across clinical trials. This is a huge step forward and will really help us.”