Supportive Care Trials Face Challenge of Scientific and Ethical Inadequacies
In a systematic review published June 29 in the Journal of Clinical Oncology, researchers led by Dr. Nathan I. Cherny from the Shaare Zedek Medical Center in Jerusalem, Israel, argued that the vast majority of clinical trials comparing an experimental treatment against best supportive care (also called palliative care) have been neither scientifically valid nor ethical, due to a lack of standardization in the supportive care provided to patients.
Studies have previously indicated that supportive care in the clinic routinely fails to provide adequate pain and other symptom management or psychological and family support, stated the authors. "In addition," they explained, "evidence indicates that oncologists frequently feel inadequately prepared for this aspect of their work." In this review of oncology clinical trials published between 1966 and 2008, these issues that bedevil supportive care delivered in the clinic have now been shown to carry over into research.
For more articles on palliative care, see the September 9, 2008 issue of the NCI Cancer Bulletin. This themed issue included the following articles:
Studies Lacking Standards
"If you're going to design a clinical trial where your control arm is supportive care, you need to embed in the protocol a standardized approach for how patients randomized into that arm will be treated, and it needs to be just as well defined as the experimental arm," said Dr. Ann O'Mara, head of palliative care research in NCI's Community Oncology and Preventive Trials Research Group. Dr. Cherny and his colleagues did not find this sort of standardized approach in the trials they reviewed.
The researchers identified clinical trials published in English between 1966 and 2008 for patients with advanced or metastatic cancers for which there were no standard treatments. These trials either compared an experimental treatment regimen and supportive care against supportive care alone (20 trials) or compared an experimental treatment regimen alone against supportive care (12 trials).
All but three of the identified trials provided minimal or no descriptions of the supportive care provided. In 21 of the studies, the supportive care provided was not standardized but instead was at the discretion of the treating investigator. Only one study described the training and experience of those providing the supportive care. "In most of the studies, the standard of [supportive care] was manifestly ad hoc, and there was no apparent use of invoked standards," stated the authors.
Implications for Study Results, Ethics
Deficiencies in the supportive care provided in these trials raise questions about the reported scientific outcomes. If patients in these trials did not receive appropriate, standardized supportive care, there's the risk "of a straw-man effect," explained Dr. Amy Abernethy, program director of the Duke University Cancer Care Research Program and one of the authors of the JCO study. "If the supportive care arm to which these new treatments are being compared is not standardized or as good as it could be, the difference between what we're calling best supportive care and the proposed new treatment can be exaggerated," she said. "I think that is the most important point we make in this article."
The lack of supportive care standards observed in these trials also raises ethical issues. "The contract you make with patients when they enroll in a trial is that you're going to make sure they have access to the best care they can have access to, the current gold standard," said Dr. Abernethy.
The authors of the JCO study recommend "that researchers and IRBs critically review [best supportive care] studies currently open," and that trials not providing this standard be amended or closed.
Reflections of Clinical Barriers
It's important to remember, said Dr. O'Mara, "that supportive care has really only taken off in the last 10 years," and that even with the recent development of guidelines and standards associated with supportive care, "a lot of people don't realize that there's a lot they don't know."
To identify and reduce the barriers to delivering appropriate symptom management and palliative care in the clinic (including a lack of knowledge among clinicians of current recommended best practices), NCI released a request for grant applications in 2004 on this subject, and the first findings from the 16 funded projects are beginning to emerge.
One major barrier that cannot be addressed with research alone, said Dr. Diane Meier, director of the Center to Advance Palliative Care and an NCI grantee, is the current reimbursement system in the United States. "Oncologists are paid to give treatment. There is no incentive to have, for example, a 90-minute family meeting about goals of care. In fact, there's a huge financial penalty. Unless we get a regulatory and reimbursement system that actually pays for outcomes, it's going to be very difficult to change this."
Reforming Best Supportive Care Studies in Oncology
"As someone who performs clinical trials, what I need is a menu of necessary supportive care that can be followed and standardized as part of the protocol, and it has to be fundamentally practical—something I can feasibly do," said Dr. Abernethy.
No such menu currently exists, but her co-author, Dr. Yousuf Zafar, recently received an NIH grant to develop standards for supportive care in oncology clinical trials. He is assembling an expert panel of oncologists, clinical trial investigators, and palliative care specialists to identify both the major problems with best supportive care delivered in the clinical trials setting and potential solutions. An independent group of clinical trial investigators will then evaluate the feasibility of the panel's recommendations.
The project's final step will be a clinical trial to compare outcomes and quality of life between patients receiving the newly defined standard of supportive care and those receiving ad hoc supportive care, as is often provided in the clinic. "This will give us a solid foundation upon which to build evidence-based supportive care" in clinical trials, explained Dr. Zafar.
— Sharon Reynolds