The Evidence for Early Palliative Care in Cancer Patients

Kate M. O'Rourke


December 02, 2015

In recent years, evidence has grown for using palliative care early in cancer patients, but it is still not incorporated as much as it should be. At the 2015 European Cancer Congress (ECC), clinicians discussed some of the challenges to integrating palliative care into oncology care.

The Evidence For Palliative Care

According to Marie Bakitas, DNSc, associate director of the Center for Palliative and Supportive Care at the University of Alabama at Birmingham, there is plenty of evidence for the benefits of early palliative care. In 1990, the World Health Organization recommended the use of early palliative care at the time of cancer diagnosis in patients with all stages of cancer, but it has taken awhile for this recommendation to be embraced.[1]

In 1990, the World Health Organization recommended the use of early palliative care at the time of cancer diagnosis in patients with all stages of cancer

In the late 1990s, the Robert Wood Johnson Foundation funded demonstration projects to test approaches to improve end-of-life care, including ENABLE (Educate, Nurture, Advise, Before Life Ends).[2] In the pilot project, patients with advanced cancer underwent four structured in-person sessions with a palliative care advanced practice nurse. The sessions focused on problem-solving, empowerment, symptom management, support and communication, and advanced care planning.

Bolstered by the success of the pilot, researchers launched ENABLE II, which was similar but involved a phone-based intervention.[3] ENABLE II was the first study to demonstrate that early palliative care improves quality of life (P=.02); mood (P=.02); and, although not statistically significant, symptom intensity (P=.06).

It wasn't until 2010, however, that the landmark study by Temel and colleagues[4] overturned the long-held belief that palliative care was only for patients with end-stage disease. This study randomly assigned 151 ambulatory patients with newly diagnosed metastatic non-small cell lung cancer to receive palliative care integrated with standard oncologic care or standard oncologic care alone.

Patients who received early palliative care, even though they received less aggressive end-of-life care, survived longer than patients receiving standard care (median, 11.6 vs 8.9 months; P=.02).[4] Patients in the early palliative care group had a better quality of life (mean score on the Functional Assessment of Cancer Therapy-Lung scale, 98.0 vs 91.5; P=.03) and were also less likely to have depressive symptoms (16% vs 38%; P=.01).[4] Individuals reporting they were seriously ill had shorter survival than those who did not consider themselves terminally ill, even after adjustment for a decline in physical and functional well-being (hazard ratio [HR], 1.50; P=.019).[5]

In 2015, results from the ENABLE III trial showed that the earlier palliative care is given, the better the outcome.[6] The trial randomly assigned 207 patients with advanced cancer to receive palliative care 30-60 days after diagnosis (early palliative care) or 3 months after diagnosis. The 1-year survival rate was drastically better among patients who received early palliative care (63% vs 48%; P=.038). There was no difference in patient-reported outcomes.

Caregivers of patients receiving early palliative care had lower rates of depression and stress.[7] Early palliative care was associated with a greater improvement in patient depression, but the hypothesis that the treatment of depression mediated the survival benefit from early palliative care was not supported.[8]

According to Dr Bakitas, the link between depression and survival is one intriguing finding from the palliative care research. In a yet-to-be published analysis of data from ENABLE II and ENABLE III, depression was associated with reduced survival (HR, 1.042; P=.001), and the palliative care intervention improved survival in depressed patients (HR, 0.963; P=.018). Patients with high depression who received usual oncologic care had the lowest survival, followed by patients who had low depression and usual care and then patients who received any palliative care intervention, early or late.

link between depression and survival is one intriguing finding from the palliative care research

So, what is the mechanism for this? According to Dr Bakitas, depressive affect could be directly associated with decreased survival and, therefore, interventions that decrease depressive affect improve survival. Epidemiologic studies indicate that stress and chronic depression are risk factors for development and progression of cancer.[9] "Biological features of immune response and neuroendocrine regulation end up not only decreasing tumor growth but also improving survival," said Dr Bakitas.

The mechanism could also be indirect, in that depression adversely affects self-care behavior, such as treatment adherence, and subpar self-care then decreases survival. Early palliative care could reduce depression and change the trajectory of self-care behaviors, resulting in improved survival, said Dr Bakitas.

A third explanation is that depression is correlated with an unknown factor, such as an unmeasured somatic symptom, that influences survival. Additional research is needed to answer the mechanism question.

Improving Education

According to many clinicians who spoke at the symposium, the evidence base for using early palliative care is solid. It's clear, however, that being in distress is not enough to push patients to ask for palliative care. In a study presented at the meeting of 169 patients with advanced cancer receiving care from 20 oncologists, the vast majority reported unmet palliative care needs, most commonly psychological or emotional distress (62%) and symptoms (62%), but the need for palliative care services did not increase the likelihood that patients would request palliative care services.[10]

According to researchers at Aurora Health Care, based in Hingham and Sheboygan, Wisconsin, patients may be reluctant to meet with additional unfamiliar providers and support teams during the final months of life and may also be concerned about time and financial constraints.[11] At Aurora, structured outpatient palliative care family conferences are improving satisfaction with end-of-life discussions.[11]

According to Michael Hoerger, PhD, an assistant professor of psychology and psychiatry at Tulane Cancer Center in New Orleans, Louisiana, one of the big stumbling blocks is the image of palliative care. "Approximately 4 out of 5 in the general public do not know what palliative care is. It is often incorrectly equated with hospice and it is often viewed as 'giving up' or 'the grim reaper,'" said Dr Hoerger. "Oftentimes, palliative care is not utilized until the very final weeks of life." In a recent survey of 349 hematologic oncologists, 55.9% reported that the typical timing of end-of-life discussions was too late.[12]

one of the big stumbling blocks is the image of palliative care

At the symposium, Dr Hoerger presented results from a recently completed pilot study that evaluated whether psychoeducational information can increase preferences for palliative care among patients with cancer.[13] The researchers randomized 598 palliative care-naive adults with cancer to receive psychoeducational information summarizing the results of Temel and colleagues' pivotal study. The summary was written in plain language and used a graphical format.

"There is a mix of complex terms and simplified language. There is also a balance of describing the results of the study and qualifying that with uncertainty," said Dr Hoerger. Participants completed measures of their health history, demographics, socioeconomic status (education level, financial strain, subjective financial standing), and severity of depression symptoms. They also completed a pre/post measure of preferences for palliative care. 

Relative to controls, participants receiving the intervention experienced an increase in preference for palliative care (P<.0001). The standardized mean difference in pre/post change was greater than a standard deviation in magnitude—a "large" effect size by conventional standards, said Dr Hoerger. Roughly one quarter of patients who received the education had a "reliable" increase in the preference for palliative care, compared with only 2% of the control group patients (23.7% vs 1.7%; P<.0001).

This "reliable increase" can be thought of as similar to a complete response endpoint in clinical trials, said Dr Hoerger. More than three quarters had an increase in preference for palliative care (75.1% vs 31.5%; P<.0001), which can be thought of as finding a partial response. Comparable effects were seen across key patient subgroups, including patients of low socioeconomic status, patients living in low-access states, and patients with more severe symptoms of depression. 

"The intervention was very effective for men but very, very effective for women," said Dr Hoerger. "Disseminating evidence for palliative care sizably increases preferences for palliative care." Dr Hoerger said he plans to scale up the intervention, incorporating stakeholder input and adding individually tailored content.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.