Professor Dame Lesley Fallowfield (University of Sussex, Brighton, UK) and Dr Ethan Basch (University of North Carolina at Chapel Hill, USA)
The reporting of drug safety in oncology clinical trials can be currently defined as “physician-centric.” But research has shown that clinicians “systematically downgrade” the severity and frequency of side effects [1,2].
By contrast, patients are more sensitive to unwanted effects and can often identify them earlier than clinicians [3], making patient-reported outcomes (PROs) important prognostic markers that could help improve patient care.
For instance, in patients with newly diagnosed intermediate-2- or high-risk myelodysplastic syndromes, patient-reported fatigue severity has been shown to be independently associated with survival [4].
Yet an analysis of 325 phase III medical oncology randomized controlled trials published between 2007 and 2011 showed that the majority (62%) did not provide information on PROs [5].
Clearly, there is a discrepancy here – PROs are useful, but are not reported routinely, which begs the question: what are the barriers to the routine assessment and reporting of PROs? medwireNews spoke to Professor Dame Lesley Fallowfield (University of Sussex, Brighton, UK) and Dr Ethan Basch (University of North Carolina at Chapel Hill, USA) to try and understand the underlying reasons for this disconnect, and to explore the changes taking place.
Breaking down the barriers
While PROs are collected in the majority of drug trials and often expected or required by regulatory authorities, Dame Lesley and Basch believe that they are not given the same weight as physician-reported safety information – frequently reported months after the primary outcome and safety data have been published, and often “in a journal with lesser prestige,” said Dame Lesley.
She pointed out that PROs are often criticized for not differing between experimental and control groups. But she attributes this to the PRO aspect of the trial not being “thought through,” often resulting in the wrong tools being used.
“What I want to know is how many patients in each treatment group achieved some benefit or some decline in their quality of life, and what the trajectory of side effects was.”
For example, “extraordinarily good tools” exist for measuring PROs in breast cancer trials, such as the breast versions of the EORTC and FACT quality of life questionnaires. But Dame Lesley stresses that they are more effective for women in the earlier stages of disease as they include questions pertaining to how the patient feels about how their breasts look after surgery. This is relevant if you are assessing a new surgical technique in patients with early-stage disease, but does not really apply to women with metastatic disease – “how their breasts look is the least of their concerns,” she said.
Furthermore, the statistical analyses of the collected data need to be done well, she told medwireNews. Just reporting a mean difference between treatment groups is not “particularly relevant,” she says, adding: “What I want to know is how many patients in each treatment group achieved some benefit or some decline in their quality of life, and what the trajectory of side effects was.”
Once again using the example of patients with metastatic disease, she explained that not only is it important to assess how much longer they live, but also how well they live during that time. If a patient maintains a steady state – that is, there is no worsening from baseline in terms of say, fatigue or pain – then that is a benefit.
Basch pointed out the potential “detrimental” effects of not reporting PROs – not just for the patients, but for the pharmaceutical companies too. PROs for pain collected during the development of abiraterone acetate for metastatic castration-resistant prostate cancer, which showed that the agent delayed pain progression relative to control [6], now form part of the drug label. Such outcomes were not collected in the pivotal trial for radium 223, however, and so while anecdotal evidence suggests that it may improve pain more than other agents, it cannot be included on the label. “It is unfortunate for the sponsor, and for the clinicians and patients,” he said.
Changing the current thinking
Most products do not have a major effect on overall survival, he highlights, “but if I know that my patients are going to feel better after taking a particular drug, then that’s much more meaningful.”
Basch highlighted that clinicians are “almost obsessed” with survival and radiographic endpoints, often forgetting that one of the most important characteristics of a drug is the effect it has on how people feel and function.
Most products do not have a major effect on overall survival, he highlights, “but if I know that my patients are going to feel better after taking a particular drug, then that’s much more meaningful.”
Dame Lesley said that clinicians have a “touching reliance and faith” that the investigator-reported CTCAE identifies all problems associated with a particular agent, “which we know is not true.” Researchers tend not to ask about things that they do not expect to see, which is why many side effects only become apparent after the trials are concluded, when the drug enters clinical practice.
Basch thinks that old-school investigators have a bias against PROs, believing them to be unreliable, a viewpoint that he does not agree with. “I am not a quality of life researcher per se, I am a clinical investigator, but I consider such patient-reported data to be important,” he said, adding that in some cases PROs are more reliable given that the inter-rater reliability of assessing objective measures, such as CAT or bone scans, is often poor.
The game is afoot
To bring PROs front and center, Basch said that ultimately the impetus has to come from the regulatory authorities. “Drug developers do what they need to do to get their drug approved,” he observes, so there needs to be an expectation that PRO data will not only be collected, but done so in a rigorous way.
But a cultural change is also needed – an understanding that without such patient-reported information, our understanding of a product is incomplete, said Basch.
Dame Lesley believes that after a long wait, “there are a few moves afoot.” People are recognizing that there are certain things that only a patient can report reliably. Obviously, a doctor is the one who can read the scans, but when it comes to pain, fatigue, appetite, the patient is the one best placed to tell you the frequency and severity of those, she said.
In the USA, clinical trials will now include the PRO-CTCAE – a companion tool to the CTCAE that captures patient-reported adverse events, Dame Lesley told us. She added that “another heartening change that has taken far too long to come about” is that the US Food and Drug Administration and the European Medicines Agency will permit pharmaceutical companies to use a quality of life or PRO label, if the data have been collected using a validated instrument.
“Money talks,” she said. “Pharmaceutical companies are waking up to the fact that although their latest product may only have a marginal benefit with regard to survival, it may have an advantage in terms of a side effect or symptom that a patient can report.”
ProtecTing patients’ quality of life
That the winds of change are blowing is evident. The clinical and patient-reported outcomes of the ProtecT trial, which assessed three approaches for the management of localized prostate cancer detected by prostate-specific antigen testing, were published simultaneously in The New England Journal of Medicine [7,8].
Basch felt that these findings “empower patients and clinicians alike,” and are crucial for decision making. The ProtecT trial did not find a significant difference in 10-year prostate cancer-specific and overall mortality among patients who underwent active surveillance, radical surgery, or radiotherapy [7]. However, the modalities varied with respect to the effect on urinary, bowel, and sexual function and associated quality of life [8].
For instance, radical prostatectomy was associated with increased urinary incontinence and worse sexual function compared with radiotherapy or active surveillance, whereas adverse bowel function was worse in the radiotherapy arm. Basch said that the PRO data are “very useful with regard to balancing the side effects of the three treatment options,” allowing practitioners to use evidence-based, rather than just descriptive or anecdotal, information to help patients who are struggling to choose the best option for them.
“In the absence of a survival difference, quality of life becomes of the utmost importance,” said Basch.
“It is essential not only to ensure the longest possible survival, but also the best possible quality of life during that time,”
And neither he nor Dame Lesley anticipate any objections from the patients themselves in terms of the routine incorporation of PROs in clinical trials.
One might think that a patient – especially someone who is ill and undergoing debilitating treatment – may not want to complete yet another form with multiple questions. But Dame Lesley said that as long as the patients know that the information will be used to improve their care, they are happy to fill the forms. Clinicians do need to listen though, because “there is no point for a patient to say that they are in pain, if their analgesic medication is then not changed,” she said.
Dame Lesley stressed the importance of educating clinicians about how valuable these instruments and the data obtained from them can be to ensure that they are providing the best care to their patients.
“It is essential not only to ensure the longest possible survival, but also the best possible quality of life during that time,” she remarked.
medwireNews is an independent medical news service provided by Springer Healthcare Limited. © Springer Healthcare Ltd; 2017