The advent of immunotherapies has created a number of interesting challenges for oncology providers. At the 2016 Palliative Care in Oncology Symposium, specialists in the field tackled these issues.
People say they love these drugs because they don’t cause side effects. That’s not true. The traditional side effects are less, but there are other toxicities we cannot ignore, such as endocrinopathies, pneumonitis, and colitis.— David Spigel, MD
Tweet this quote
“There is a lot of newness to how we approach patient care with immunotherapies on board,” said David Spigel, MD, Director of the Lung Cancer Translational Research Program at Sarah Cannon Cancer Center, Nashville.1 “There are new expectations and new ways to practice. Clinical research has also been affected with this wave of new treatments already in the clinic.”
Jennifer Temel, MD, Clinical Director of Thoracic Oncology at Massachusetts General Hospital, Boston, said one of the biggest challenges is prognosis.2 “It’s always hard for us to disclose a poor prognosis, but especially when we feel uncertain about it,” she shared.
Four immune checkpoint inhibitors have been approved by the U.S. Food and Drug Administration for use in six different tumor types. New agents are emerging at a fast clip, and indications for current agents are expanding. Predicted soon to join ipilimumab (Yervoy), pembrolizumab (Keytruda), nivolumab (Opdivo), and atezolizumab (Tecentriq) are durvalumab, avelumab, and tremelimumab. These and others are being tested in more than 17 different cancers.
How to Use Them and in Whom
One of the challenges with immunotherapy is patient selection. In patients with non–small cell lung cancer, programmed cell death ligand 1 (PD-L1) serves as a biomarker for response to pembrolizumab. Patients whose tumors express high levels of PD-L1 are most likely to benefit from pembrolizumab. Other factors may eventually prove influential—including mutation burden, mismatch-repair defect, and some clinical characteristics—but this remains unproven.
High expression of PD-L1 (≥ 50%) is required for pembrolizumab’s use, and this is “problematic,” mainly because of variability and different cutoffs among the available assays, according to Dr. Spigel. For other anti–PD-1/PD-L1 agents, the level of expression may or may not be important. “The data are mixed,” he admitted.
“We are starting to see evidence that pembrolizumab may be the only drug where PD-L1 expression is necessary for its use…. Currently, we don’t have a way to use these assays in routine patient care,” he said.
Management of unfamiliar toxicities creates another challenge. “People say they love these drugs because they don’t cause side effects. That’s not true,” declared Dr. Spigel. “The traditional side effects are less, but there are other toxicities we cannot ignore, such as endocrinopathies, pneumonitis, and colitis. The trend to combine checkpoint inhibitors worries me even more, as serious adverse events occur in 40% of patients receiving an anti–PD-1 agent plus ipilimumab.”
In the clinic, more time is spent on patient education, the clinic flow is different (fewer patients in infusion chairs), and follow-up/monitoring has changed as well, since scans and RECIST (Response Evaluation Criteria in Solid Tumors) criteria are less helpful, he added.
Less Incentive for Clinical Trial Enrollment
Another challenge pertains to clinical trial enrollment, since patients can often obtain access to immunotherapies off label. “It’s hard to convince patients to go onto clinical trials, or even to convince physicians, when it’s easy to give nivo or pembro for whatever tumor now,” Dr. Spigel stated. Clinical trial enrollment is critical for testing the wealth of drugs now in development that may hold even greater promise, including LAG3 and TIM3 inhibitors, costimulatory agents and other immune modulators, CAR T cells, and vaccines. They must be tested alone and in combinations and sequences, which certainly strains resources, he added.
It’s incredibly hard to sit across from another human being and the people who love them and tell them they have a life-threatening illness. And it’s even harder for us now because there’s so much optimism in the oncology field.— Jennifer Temel, MD
Tweet this quote
“We have approvals across multiple cancer settings, with more coming, and this means new challenges for patients and providers,” Dr. Spigel said.
Prognostication Is More Difficult
Dr. Temel discussed what she called “the changing conversation around prognostication” for patients on immunotherapies. “These conversations have always been difficult. It’s incredibly hard to sit across from another human being and the people who love them and tell them they have a life-threatening illness,” she said. “And it’s even harder for us now because there’s so much optimism in the oncology field.”
Exactly which patients will be responders, however, is never clear at diagnosis, and this makes prognostication especially challenging. “It’s the uncertainty and inability for us to feel confident about what will happen to an individual patient in the future that make these conversations so very difficult now,” she said.
Dr. Temel drew from her own experience as a lung cancer specialist, noting that in the early 2000s, four regimens delivered virtually the same outcomes in metastatic disease, and one could predict where a patient would fall along the Kaplan-Meier curve. Today, that is much more difficult.
She illustrated this difficulty by describing a recent 47-year-old patient who, after having Hodgkin lymphoma when younger, developed a neuroendocrine tumor in the lung that eventually metastasized to the brain, leaving her with pain, anorexia, and cachexia. Dr. Temel fought to enroll her on a trial of an anti-PD-1 agent (nivolumab had yet to be approved), but she was excluded on the basis of radiation-related second malignancies. With the patient “actively dying,” the insurance company “miraculously” agreed to nivolumab, and she became “one of the incredibly fortunate patients” to have a complete response. This patient just returned from a family vacation to Italy and “is doing fabulously,” reported Dr. Temel. “We had prepared for her death, but she had one of those exceptional responses.”
These patients are difficult to predict, and when oncologists get it wrong, it’s often to err on the side of optimism. A prospective study of 468 terminally ill patients and their 343 oncologists found that only 20% of physicians’ estimates of prognosis were accurate; 63% were overly optimistic.3
“We too must have prognostic awareness. We are hopeful about the state of science in oncology and want our patients to maintain hope,” said Dr. Temel, adding that physicians “don’t want to give patients the wrong idea” about prognosis. “We have all had conversations with family members who said, ‘They told us Dad would be dead in 3 months, and he’s alive 1 year later.’ And most don’t say this in a grateful way. They’re upset!”
Why Prognostication Is Important
These conversations are troublesome but important, she continued. When patients understand their prognosis and the likely trajectory of their illness—ie, have “prognostic awareness”—they make more appropriate decisions about their medical care and can determine how best to spend the time they have left. They are also more likely to engage in conversations about end-of-life care.
Virtually all studies have concluded that patients and their families do want at least a sense of the future. But despite this acceptance, “our practice is not to give them this information,” Dr. Temel indicated. Most oncologists provide prognostic information only upon request. Furthermore, when they do, patients and families often do not understand it. “Studies have shown that most patients and their families do not know that metastatic disease means a cancer is incurable,” she explained. “We must make sure that patients understand their disease.”
Ongoing Discussions About Prognosis
These conversations require time and space and can be hard to squeeze into a 15-minute appointment. Prognosis should be discussed “at a time when the patient can hear it, with language the patient can understand, and with compassionate communication,” suggested Dr. Temel. Honesty is the best policy here, she added, but indicated the conversation can allow for hope.
The aim is to acknowledge the prognostic uncertainty but frame it in the context of new scientific discoveries. The strategy she prefers is like a “best case, worst case scenario,” although she does not use that phrase. She prefers to say, “Ten years ago, we hardly ever saw patients with metastatic non–small cell lung cancer live more than 1 year, but with all our latest discoveries, sometimes that happens. A proportion of metastatic lung cancer patients treated with nivolumab are living for years, and we don’t know if you will be one of them or not.”
These conversations should not be “one-time events.” Prognosis should be introduced upon diagnosis and then “woven into the course of care,” she suggested. “At diagnosis, I disclose the incurable nature of their illness, and then throughout the illness, I refine the prognosis and address the goals of care. When there is significant decline, we transition to focusing on comfort.”
The impact of a changing prognosis, and the associated conversations, can be tough on patients, Dr. Temel admitted, relaying a story about a woman who underwent a lung biopsy at each disease progression. Each time, rapid tumor growth would be halted. Each time, she had prepared to die, only to “awake from the dead,” and the cycle continued. Ultimately, the patient refused a final targeted treatment to which she might have responded. She said she could not continue to say goodbye to her family, “just to come back and have to face it all again.”
“We need to figure out how best to support our patients and their families,” Dr. Temel concluded. “Our field is changing rapidly, and it’s having an impact on how patients deal with cancer day to day.” ■
Disclosure: Drs. Spigel and Temel reported no potential conflicts of interest.
1. Spigel DR: Immunotherapeutics: Changes in the landscape of advanced cancer care. 2016 Palliative Care in Oncology Symposium. General Session 1. Presented September 9, 2016.
2. Temel JS: The changing conversation around prognostication. 2016 Palliative Care in Oncology Symposium. General Session 1. Presented September 9, 2016.
3. Christakis NA, Lamont EB: Extent and determinants of error in doctors’ prognoses in terminally ill patients: Prospective cohort study. BMJ 320:469-472, 2000.