Terminal cancer patients who receive chemotherapy in the last months of their lives are less likely to die where they want and are more likely to undergo invasive medical procedures than those who do not receive chemotherapy, according to research in this week's BMJ.
The findings underscore a disconnect between the type of care many cancer patients say they want and the kind they receive, and highlight the need for clearer and more balanced discussion of the harms and benefits of palliative chemotherapy at the end of life by doctors, patients and families, the study authors say.
The Dana-Farber Cancer Institute and Weill Cornell Medical College study found sobering outcomes for patients who received palliative chemotherapy -- treatment designed to prolong survival and ease symptoms, but not to cure disease -- among patients whom physicians had determined had six or fewer months to live. Whereas 80 percent of patients who did not receive palliative chemotherapy died where they wished, only 68 percent of those whose disease management included palliative chemotherapy died in the place they wanted to. Nearly 66 percent of patients who did not receive palliative chemotherapy died at home, compared to 47 percent of patients who received palliative chemotherapy. And patients who received palliative chemotherapy were much more likely than their counterparts to die in an intensive-care unit -- a contrast of 11 to 2 percent.
"Our finding that patients with terminal cancers were at higher risk of receiving intensive end-of-life care if they were treated with palliative chemotherapy months earlier underscores the importance of oncologists asking patients about their end-of-life wishes," said Alexi Wright, MD, MPH, an assistant professor and medical oncologist at Dana-Farber Cancer Institute and lead author on the study. "We often wait until patients stop chemotherapy before asking them about where and how they want to die, but this study shows we need to ask patients about their preferences while they are receiving chemotherapy to ensure they receive the kind of care they want near death."
"It's hard to see in these data much of a silver lining to palliative chemotherapy for patients in the terminal stage of their cancer," said senior author Holly Prigerson, PhD, the Irving Sherwood Wright Professor in Geriatrics and Palliative Medicine at Weill Cornell Medical College. "Until now, there hasn't been evidence of harmful effects of palliative chemotherapy in the last few months of life. This study is a first step in providing evidence that specifically demonstrates what negative outcomes may result. Additional studies are needed to confirm these troubling findings."
The reasons for the link are complex, but may originate in patients' misunderstanding of the purpose and likely consequences of palliative chemotherapy and lack of acknowledgment of their own prognoses, said Prigerson, who recently came to Weill Cornell from Dana-Farber Cancer Institute in Boston. In the study, patients getting palliative chemotherapy were less likely to talk to their oncologists about the care they wanted to receive if they were dying, to complete Do-Not-Resuscitate orders, or to acknowledge they were terminally ill, she noted. With 56 percent of patients receiving palliative chemotherapy in their final months, the findings underscore the potential harms of aggressive use of chemotherapy in dying patients, and the possible need for widespread changes in oncology practice at academic medical centers.
The scientists analyzed data from 386 patients in the federally funded Coping with Cancer study, which followed terminally ill people and their caregivers until after the patients died. During the six-year study, researchers examined how psychosocial factors influenced patients' care. In the month after the patients died, caregivers were asked to rate their loved ones' care, quality of life, and place of death as being where the patient would have wanted to die. The investigators then reviewed patients' medical charts to determine the type of care they actually received in their last week.
Even after the researchers took into account characteristics such as age, marital status, whether a patient had health insurance, their overall physical and mental health, and the treatment they preferred, those who received palliative chemotherapy still underwent more invasive medical procedures in the last week of their lives and in more cases died in ICUs than patients who did not receive palliative chemotherapy. Among those interventions were CPR and mechanical ventilation.
Patients who received palliative chemotherapy were also more likely to be referred to hospice care, which provides comfort care and emotional support, a week or less before they died. Some 54 percent of patients receiving palliative chemotherapy were referred to hospice late, compared to about 37 percent of patients who weren't taking the drugs.
"The results highlight the need for more effective communication by doctors of terminal prognoses and the likely outcomes of chemotherapy for these patients," said Prigerson, who was the principal investigator on the Coping with Cancer study. "For patients to make informed choices about their care, they need to know if they are incurable and understand what their life expectancy is, that palliative chemotherapy is not intended to cure them, that it may not appreciably prolong their life, and that it may result in the receipt of very aggressive, life-prolonging care at the expense of their quality of life."
The results should not be taken to mean that patients should be denied or not offered palliative chemotherapy, however. "The vast majority of patients in this study wanted palliative chemotherapy if it might increase their survival by as little as a week," added Wright. "This study is a step towards understanding some of the human costs and benefits of palliative chemotherapy."
- A. A. Wright, B. Zhang, N. L. Keating, J. C. Weeks, H. G. Prigerson. Associations between palliative chemotherapy and adult cancer patients' end of life care and place of death: prospective cohort study. BMJ, 2014; 348 (mar04 4): g1219 DOI: 10.1136/bmj.g1219
Cite This Page: