Clinician perspectives on integrating neuro-oncology and palliative care for patients with high-grade glioma

Abstract

Background

Patients with high-grade glioma have high palliative care needs, yet few receive palliative care consultation. This study aims to explore themes on (1) benefits of primary (delivered by neuro-oncologists) and specialty palliative care (SPC) and (2) barriers to SPC referral, according to a diverse sample of clinicians.

Methods

From September 2021 to May 2023, 10 palliative physicians and 10 neuro-oncologists were recruited via purposive sampling for diversity in geographic setting, seniority, and practice structure. Semistructured, 45-minute interviews were audio-recorded, professionally transcribed, and coded by 2 investigators. A qualitative, phenomenological approach to thematic analysis was used.

Results

Regarding primary palliative care, (1) neuro-oncologists have primary ownership of cancer-directed treatment and palliative management and (2) the neuro-oncology clinic is glioma patients’ medical home. Regarding SPC, (1) palliative specialists’ approach is beneficial even without disease-specific expertise; (2) palliative specialists have time to comprehensively address palliative needs; and (3) earlier SPC enhances its benefits. For referral barriers, (1) appointment burden can be mitigated with telehealth, home-based, and embedded palliative care; (2) heightened stigma associating SPC with hospice in a population with high death anxiety can be mitigated with earlier referral to promote rapport-building; and (3) lack of neuro-oncologic expertise among palliative specialists can be mitigated by emphasizing their role in managing nonneurologic symptoms, coping support, and anticipatory guidance.

Conclusions

These themes emphasize the central role of neuro-oncologists in addressing palliative care needs in glioma, without obviating the need for or benefits of SPC. Tailored models may be needed to optimize the balance of primary and specialty palliative care in glioma.

Palliative care is defined as “active holistic care of individuals across all ages with serious health-related suffering due to severe illness and especially of those near the end of life…to improve the quality of life of patients, their families, and their caregivers.”¹ To achieve this, palliative care addresses 8 clinical domains as defined by the National Hospice and Palliative Care Organization (NHPCO): structures and processes of care (ie, interdisciplinary care and developing goal-aligned care plans); physical symptom management; psychological symptom management; psychosocial support; spiritual support; culturally competent care; end-of-life care; and ethical and legal aspects of care (ie, advance care planning).² Clinical trials in systemic cancer and other serious illnesses have shown that receipt of palliative care from specialized teams yields improvements in quality of life, reduced symptoms, greater patient and family satisfaction with care, and greater alignment of treatment with a patient’s goals and values.³ Indeed, the American Society of Clinical Oncology and the National Comprehensive Cancer Network recommend early palliative care (within 2–3 months of diagnosis) for all patients with aggressive cancer.^4,5 This recommendation applies to high-grade (WHO Grades 3 and 4) glioma (HGG) and is an explicit recommendation of the European Association for Neuro-Oncology, who acknowledge that there is limited evidence in this specific population.^6,7

Patients with HGG also have significant needs across the domains of palliative care.⁸ Physical and psychiatric symptoms and existential distress are prevalent from early in the disease course.^9,10 Early cognitive impairment enhances the importance of goals of care discussions near the time of diagnosis, while a prolonged period of functional decline at the end of life adds complexity to navigating end-of-life care.^11,12 Despite these needs, only a minority (15%–40%) of HGG patients are ever referred to palliative care.^13,14 When referrals are made, they typically occur later than the guideline recommends: in a study of more than 10 000 Medicare recipients with glioblastoma, only 6.25% were referred to palliative care within 10 weeks of diagnosis (an additional 9% were referred later in the disease course)¹⁵; another retrospective analysis of 621 patients from a multihospital academic cancer registry found that among 134 (22.6%) patients ever referred to palliative care, nearly half of the 134 were referred at the end of life after discontinuing cancer-directed treatment. The reasons for this gap in care are unclear.

Previous studies have identified barriers to early palliative care referral in other systemic cancers; these barriers include: oncologists’ and patients’ lack of awareness of the role of palliative care; difficulty coordinating care; emotional responses among oncologists and patients; and stigma about palliative care.¹⁶ While some or all of these barriers likely also exist for HGG, there is an important clinical difference that warrants investigation: HGG is a neurologic disease, which is not a focus of palliative medicine training. Additionally, the majority of palliative medicine specialists completed residency training in internal or family medicine, not neurology.¹⁷ Where refractory symptoms (eg, pain, nausea, and depression) are indications for specialty palliative medicine referral in systemic cancers, the symptoms typical among patients with HGG and other neurologic diseases (eg, seizures, change in cognition, impaired mobility, and headache) fall outside of the expertise of most palliative medicine specialists.¹⁸ As a result, there may be a greater need for primary palliative care for basic symptom management and goals of care discussions in HGG, which are delivered by neuro-oncologists and other nonpalliative clinicians.¹⁸ Thus, the actions/practices of neuro-oncologists and palliative medicine specialists, respectively, in meeting the palliative care needs of HGG patients should be elucidated. This qualitative study explores themes on the benefits of primary and specialty palliative care, as well as strategies to improve palliative care referral rates, from the perspectives of a diverse sample of neuro-oncologists and palliative medicine specialists.

Methods

Design and Setting

Using a phenomenological approach, a minimum sample size of 20 clinicians (10 neuro-oncologists and 10 palliative medicine specialists) was assessed as being likely to capture the full range of physician perspectives on the study topic.¹⁹ Neuro-oncologists and palliative medicine specialists were eligible to participate if they were board-certified in their specialty and in active practice caring for adult patients. Potential participants were identified via purposive sampling with the goal of including (1) clinicians with a high level of experience in both specialties (ie, neuro-oncologists with training/interest in palliative care and palliative specialists with known experience with neuro-oncology) and (2) roughly equal distribution of sex, geographic location across the United States, and years in practice. To recruit clinicians with a high level of experience in both specialties, an initial 9 individuals (4 neuro-oncologists and 5 palliative medicine specialists) known by study team members to meet these criteria were approached; 2 palliative medicine specialists did not respond, and the remaining 7 physicians (4 neuro-oncologists and 3 palliative medicine specialists) were recruited and interviewed.

For further recruitment, an IRB-approved post inviting potential participants to contact the study team was shared with the neuro-oncology special interest group listserv of the American Academy of Neurology (by J.W.T.), but it did not yield any respondents. The American Academy of Hospice and Palliative Medicine does not permit this type of post.

Thus, to complete recruitment of 10 neuro-oncologists according to our rubric of the above-named characteristics, J.W.T. referred 6 additional neuro-oncologists known to her (but not known to the remainder of the study team), and we recruited all of them. R.C.C. requested and received 9 referrals for palliative medicine specialists (not known to her or the study team) from colleagues at other institutions; of these, 7 were recruited, enabling us to achieve our target of 10 palliative medicine specialists.

Overall, a total of 24 clinicians were approached: 4 (all palliative medicine specialists) did not respond or were unavailable. The remaining 20 were enrolled to make up the final sample, which met our goals for the above-named range of certain characteristics.

Data Collection

We developed 2 semi-structured interview guides (1 for neuro-oncologists and 1 for palliative medicine specialists; both are available as Supplementary Materials) organized around the 8 NHPCO palliative care domains. Within each domain, participants were asked about the types of care delivery they perform for patients with HGG, as well as their perspectives on the types of care delivery they expect the other type of specialist to perform. The structure and content covered were similar for both specialties, but specific questions differed to account for the role of palliative specialists as consultant. For example, for the “structures and processes of care” domain, neuro-oncologists were asked, “How do you approach goals of care discussions with HGG patients and families in your practice? What role do palliative care specialists play, if any?” For the same domain, palliative medicine specialists were asked, “Are there any challenges to goals of care discussions for HGG patients compared to other patients you might see? What aspects do you think are primarily the responsibility of the neuro-oncologist and what aspects do you feel most comfortable with?” Aside from the domain-specific questions, which were the primary focus of the current manuscript, the neuro-oncology interview guide also included a section about experiences with specialty palliative care (how often and why they refer, positive and negative experiences with referral, and how they collaborate with palliative specialists when patients are referred). The palliative specialist interview guide also included a section about unique challenges in caring for patients with HGG, differences between HGG and other types of serious illness, and positive and negative experiences caring for patients with HGG. Participants from both groups were also asked about their training and clinical experience, as well as the structure of their practice. Details of practice structure included whether outpatient specialty palliative care is delivered via stand-alone clinic (referral to a separate clinic without specific affiliation with neuro-oncology) or via embedded clinic (one or more palliative specialists are specifically affiliated with neuro-oncology, have dedicated time for seeing neuro-oncology patients, and may or may not be colocated with neuro-oncology).²⁰ Demographic data such as provider age, race, and ethnicity were not collected to minimize question burden and because these factors were felt to have limited relevance to the scope of the study.

Interviews were conducted via HIPAA-compliant Zoom. Informed consent was obtained at the beginning of the interview. Nineteen of 20 interviews were conducted by a medical sociologist (K.G.) not known to any of the participants. The first interview was also observed by R.C.C., and 1 neuro-oncologist interview was conducted by R.C.C. due to a scheduling conflict. Each interview lasted approximately 45 minutes. Participants were compensated $50 upon completion of the interview. All interviews were recorded and professionally transcribed.

Data Analysis

De-identified transcripts were imported into MAXQDA 2022 (VERBI Software, 2022) for analysis. Two investigators (R.C.C., a neurologist with fellowship training and board certification in hospice and palliative and medicine, and J.F.N., an internal medicine physician and doctoral candidate) independently open-coded the first 2 transcripts. All coinvestigators then reviewed and compared the resulting codes and met to jointly develop a preliminary codebook. This preliminary codebook was applied (by R.C.C. and J.F.N.) to a third transcript and additional codes were added using an iterative, comparative method.²¹ The results were discussed with the teams’ qualitative content expert (K.G.) to develop a final codebook. R.C.C. and J.F.N. then used this code book to analyze the remainder of the transcripts as interviews were completed, meeting weekly to compare results and resolve discrepancies. When the target minimum enrollment was reached, the investigator team met to identify and discuss emergent themes. All transcripts from both specialties were analyzed together and predominant themes are reported for the total sample, except where it is specifically noted in the Results section that there was a discrepancy between neuro-oncologists and palliative medicine specialists. Thematic saturation had been reached and thus no further participants were enrolled.

Standard Protocol Approval

This study was reviewed and approved by the Institutional Review Board at the Icahn School of Medicine at Mount Sinai (STUDY-21-01014).

Results

Participants

Participant characteristics are detailed in Table 1. Among both groups, the majority practiced in academic settings; in each group, 3 participants were located on the West Coast, 3 on the East Coast, and the remainder in other locations. Among neuro-oncologists, 2 had completed palliative medicine fellowship. Among palliative specialists, 3 had experience working in embedded clinics in neuro-oncology. None of the neuro-oncologists worked in clinics with an embedded palliative specialist.

Table 1.

Characteristics of Participants and Their Practice Structure

	Clinicians (N = 20)
Primary specialty
Neuro-oncology	10
Palliative medicine	10
Sex
Male	9
Female	6
Geographic location in US
West coast	6
Southwest	1
Midwest	3
South	2
Northeast	6
Years in practice
0–5	9
5–10	6
10+	5
Practice setting
Academic	17
Community	3
Palliative medicine fellowship
Yes	12
No	8
Experience with embedded palliative
specialist in neuro-oncology clinic
Yes	3
No	17

Primary Palliative Care

Two predominant themes were identified with respect to primary palliative care, both centering the neuro-oncology clinic at the heart of HGG management, whether for cancer-directed treatment or for addressing palliative care needs. Quotes relevant to these themes are in Table 2 and numbered in brackets in this text to correspond to supporting quotes in Table 2.

Table 2.

Themes Relating to Primary and Specialty Palliative Care With Supporting Quotes

Theme	Supporting Quotes
Primary palliative care
Neuro-oncology physicians have ultimate ownership of cancer-directed treatment and addressing palliative care needs.	[1] “[Patients] really want to get the most out of their oncology visit. They want to review scans and talk about all the treatment options. They’re really jam-packed. And often, they’re kind of a bit of a whirlwind…the oncologist usually has the longest relationship with patients and families because they’re meeting them at the beginning.”—SPC #6, neurology, embedded clinic experience* [2] “[I]n a field where most of the diseases are noncurable, we sort of conceptualize ourselves as hyperaggressive palliative care physicians because we are leading patients through that journey from the moment that we meet them. Everyone knows the endpoint. I think we’re pretty explicit to patients and their families what the endpoint is from the moment that we meet them. So, we set realistic expectations. Yet, at the same time, we employ tools that are more aggressive than most oftentimes associated with palliative care…brain surgery, radiation, chemotherapy, etc.”—NO #10, neurology, no SPC fellowship [3] “There’s a unique nature of this shared relationship of being another cog in the family around the loved one who’s dying—it forms a semi-permanent bond and it’s hard for people to give that up, especially when they see that they’re losing their loved one. So, I don’t think that that’s a role that anyone else can fill. I think, in a sense, I’m just a shoulder to cry on and a witness to what they’re going through that is not something that another physician could stand in and provide.”—NO #6, neurology, no SPC fellowship
The neuro-oncology clinic is HGG patients’ medical home.	[4] “The [oncology] nurses are actually really good at [advance care planning], so this is almost kind of one their jobs because…when they speak to our patients and families…the stakes are a little bit lower, so they feel a bit more relaxed, but they do…[address] caregiver fatigue, stress, burnout. I haven’t referred specifically, but when palliative care’s already part of the team, they are pretty helpful with that.”—NO #1, neurology, no SPC training [5] “[O]ur social workers are really excellent and very interested in following these patients and even doing some therapy, so a lot of psychosocial needs—I think I couldn’t do my job without them at all, even from a palliative care or neuro-oncology perspective.”—NO #9, neurology, SPC fellowship [6] “I work a lot with one of our chaplains, who is very interested and dedicated to this. And so, when those kinda spiritual needs of care, and the spiritual impact of care, or treatment, or cancer impacts them, then I would tend to involve chaplaincy…they have a formal kind of consult process, where they would come and explore the kind of spiritual aspects, and impacts, and influences of care with the patient.”—NO #2, neurology, no SPC fellowship [7] “[T]here’s the neuro-onc supportive care services role basket of services…if I come to that team which includes nurses, social workers, cognitive therapists, psycho-oncologists, and say I need help with this patient, they’ll be like okay here’s the resources you do it. And I’m like, no that’s actually your job…I do palliative care, I’m not a social worker. And so, I think there’s still this delineation of roles that continues to be a challenge…”—SPC #3, internal medicine, embedded clinic experience
Specialty palliative care
The palliative care approach has value beyond addressing specific needs or symptoms.	[8] “I do think it ties into the psychological, emotional, etc. I think all of that is part and parcel of it. And I think palliative care has a good mindset when they’re conceptualizing everything, right? They know how to deal with these things, the bad constipation, the bed sores, etc., which are all diminishing quality of life, which are making life harder for the patient’s family, etc. And then knowing [how to couch it] in the greater context of the individual, psychosocial, economic framework.”—NO #10, neurology, no SPC fellowship [9] “[P]atients often feel that the absence of a clear discussion about prognosis is not helpful to them…You have to invite it, you have to be delicate, and you have to be respectful of where the oncologist is. But patients would very much value clear discussions…about, ‘This is what this course typically looks like, and here’s what the roadmap ahead may look like, and here’s a sign that you’re declining, here’s a sign that you’re stabilizing, and those uncommon situations, here’s a sign that you’re actually improving.’ So, the value that we bring to this is really the education of the family, and the patient and answering their questions, and the clarity of the discussion. They don’t feel that there’s as much obfuscation.”—SPC #10, internal medicine, no embedded experience [10] “[T]he palliative care team is great. They’re willing to almost pick up the mess in which we say ‘Hey, can you have a goals and care conversation?’ and they’re very, very good at that in kind of being realistic and also empathetic at the same time. So, recently, a couple cases of an inpatient service in which the outpatient oncologist was kind of hesitant on…broaching that conversation, the palliative care team came in and very expertly navigated it in terms of—at least made the expectations to the family much more reasonable compared to what they were expecting.”—NO #2, neurology, no SPC fellowship [11] “I sent a patient to palliative care, and that patient felt very comfortable since to find a palliative care doctor wanting to stop their treatment, whereas they wouldn’t have said that to me because we’re in this cancer/therapeutic relationship where I think that they are afraid of letting me down. Something like that. So, I think if I’m in sort of a complicated family, or psychosocial situation, or that feeling that I’m not doing a very good job, or there might be something else that needs to be addressed that the patient is not really conveying.”—NO #5, neurology, SPC fellowship
Palliative care teams have time to address palliative care issues that may be eclipsed by cancer-directed care during neuro-oncology visits.	[12] “We do a ton of palliative care just in my clinic, and when I tend to refer patients to our palliative care team, it tends to be in the situations where the frequency of needs is more than I can provide in my clinic. Where the depth of coverage, the depth of discussion, the depth of the need is greater than I have the ability or time—really time—to address in the visit, and so, that’s when I tend to make those referrals. Or if I anticipate that we’re gonna get into a tight spot and a tough situation and want them on board kind of early to help to forecast challenges that may come up.”—NO #2, neurology, no SPC fellowship [13] “When it goes really, really well, what is nice about having the palliative care team is that it allows me the opportunity to just focus on the medical issues related to the patient without doing them the disservice of not attending to the psychosocial issues and the emotional part. Sometimes it’s really nice to break those up and to have a partner who can just manage that and you can just manage this and that way, I think—I think the patients will really benefit because you’re not—this sounds weird—you’re not sort of skimping on either one of the aspects of care.”—NO #8, neurology, no SPC fellowship
The role of specialty palliative care evolves throughout the disease trajectory.	[14] “[O]ne situation where I think it went well, was the patient came to rehab for a very specific goal, which was to try to improve function enough to ease the caregiving burden on family members. The goal was to get back home. The goal was to follow-up with neuro-oncology and continue treatment. I think there was maybe radiation treatment, was coming up. So, it went well in that we were able to clearly identify the goals of the patient, advocate for the patient’s goals with the rehab team, and he was actually able to show some signs of functional improvement. So that he was satisfied, family was satisfied, and he was able to get home.”—SPC #8, internal medicine, no embedded clinic experience [15] “And in terms of what I did during those visits, with any palliative care patient that I’m seeing in clinic—really the first, even second visit—is primarily focused on any urgent symptoms that they want to address. And then, really just rapport building and understanding their life at home, really trying to get a good sense of their functional status, and then also asking about their understanding of their illness. So that’s most of what I do typically the first visit or two, and then I think as time goes on—makes sense—symptom management kinda carries on throughout and may pick up towards the end of life…symptom management was one of the top three things we did in every visit for 70% of the time.”—SPC #9, internal medicine, embedded clinic experience [16] “For me, what’s rewarding are the patients where you can have a really clear conversation about what to expect as they get closer to the end of life and maybe make a recommendation for hospice or maybe address some fears that they have…patients can say, ‘I’m terrified that I’ll die gasping for air or in a lot of pain.’ And to be able to actually explain to people how we treat those things and how we prevent that suffering at the end of life, can be helpful.”—SPC #3, internal medicine, no embedded clinic experience

NO = neuro-oncology; SPC = specialty palliative care.

Quotes are numbered according to their appearance in the text.

^*Quotes are attributed by participant number (randomly assigned during the analysis phase to maintain privacy) and residency training. Neuro-oncologists are further characterized by whether or not they completed specialty palliative care fellowship, and palliative specialists are characterized by whether or not they have embedded clinic experience.

The first theme concerns the role of neuro-oncology physicians: neuro-oncologists were described as having ultimate “ownership” or principal care of HGG patients, from identifying and responding to oncologic and palliative care needs to coordinating care with other consultants (including palliative medicine specialists). One driver of ownership is the neuro-oncologist’s unique expertise, which encompasses cancer-directed therapies and clinical trials, management of neurologic symptoms and side effects of cancer-directed therapies (including titration of steroids), and prognostication [1]. The second driver of ownership is the neuro-oncologist’s relationship with HGG patients and their caregivers, which is perhaps the most consistent out of many clinicians encountered from the time of diagnosis to death. They are a source of hope in the face of incurable, rapidly progressive illness, yet this dynamic also creates a sense of obligation to be the primary source of support when hope for prolonged survival fades [2]. Thus, referral to palliative care—particularly for advance care planning and goals of care discussions—may feel like a betrayal of that obligation [3].

The second theme concerns the position of the neuro-oncology clinic as a multidisciplinary medical home for HGG patients. Neuro-oncology participants most commonly had advanced practice providers (APPs), nurses and social workers who add to the number of palliative care domains addressed in their practice. APPs and nurses may be the first point of contact for patients and caregivers who call the clinic, addressing symptoms, care coordination, and coping [4]. Social workers may address emotional and existential distress, caregiver support, practical psychosocial concerns, and completion of advance directives [5]. Chaplains, support groups for patients and caregivers, and other staff may also contribute to primary palliative care [6, 7]. For some symptoms (headaches, seizures, complex neuropsychiatric symptoms), the neuro-oncology clinic may also coordinate with other clinicians (headache specialists, epileptologists, psychiatrists).

Specialty Palliative Care

Three themes were identified concerning the perceived role(s) of specialty palliative care in HGG: (1) the holistic approach palliative medicine specialists take to patient care does not necessarily require disease-specific expertise and has inherent value that transcends the domains of palliative care; (2) palliative medicine specialists have time to address issues outside of cancer-directed therapy; and (3) earlier referral allows palliative specialists to meet the evolving needs of HGG patients throughout the illness trajectory.

With respect to the inherent value of the palliative medicine holistic approach, participants acknowledged that palliative medicine specialists generally do not have expertise or training in managing serious neurologic illness such as HGG. Management of neurologic and neuropsychiatric symptoms, motor and cognitive deficits, and prognostication were identified as the purview of neuro-oncologists or other consultants (eg, epileptologists, headache specialists, psychiatrists, and rehab specialists). However, across domains, several aspects of palliative medicine expertise were identified as beneficial: providing anticipatory guidance about symptoms and caregiving; communicating about difficult topics (eg, disease progression and end-of-life care) based on the expert assessment and endorsement of neuro-oncologists; creating space for patients to share distress and discuss coping; and extending support to caregivers and families in addition to patients [8–10]. Building relationships quickly and skillfully was also a strength, including related to difficult topics such as end-of-life care. Indeed, whereas neuro-oncologists may feel a sense of obligation and ownership surrounding these areas, the lack of a preexisting relationship with patients may allow for more open discussions on the part of both patient and clinician [11].

Secondly, palliative medicine was recognized as beneficial for patients whose needs require more time than neuro-oncology is able to dedicate, given the priority of managing cancer-directed therapy. Time-based indications for referral include high symptom burden, particularly for non-neurologic symptoms, complicated family dynamics or discordance between the patient/family and team, and complex psychosocial issues [12, 13].

Finally, the potential role of specialty palliative care evolves over the disease trajectory and depends on care setting and clinical need. Notably, neuro-oncologists primarily emphasized the role of specialty palliative care at the end of life and in cases where clinical need exceeds what is possible given time constraints. Reasons given include primary palliative care being adequate for the majority of patients, as well as patients potentially having negative perceptions of palliative care and being distressed that it was recommended. Palliative specialists favored earlier referral in general for patients with serious illness, though few had experience with earlier referral in HGG. Thus, specific examples of earlier referral (ie, prior to the end-of-life stage) as described below are drawn from palliative specialists with care structures that promote more frequent contact with HGG patients.

For the numerous patients who require acute inpatient rehabilitation following diagnosis and tumor resection, palliative specialists may assist with nonneurologic symptoms, reframing expectations around functional outcomes, and even coordinating visits with neuro-oncologists who might not otherwise see them until they are discharged [14]. During cancer-directed therapy, there may be opportunities to manage nonneurologic symptoms, build rapport with both patients and caregivers, and engage in longitudinal conversations about advance care planning [15]. At the end of life, they have expertise in coordinating care transitions (ie, facilitating discussions with patients and/or caregivers, as well as developing an appropriate care plan) [16]. In states where medical aid in dying is legal, this may include navigating patient requests for this option. For patients who are actively dying in the hospital, they may also primarily manage end-of-life care.

Barriers to Specialty Palliative Care Referral and Strategies to Overcome Them

Given the disparity between the perceived benefits of specialty palliative care and the low referral rates in HGG, as well as the contrasting perspectives of neuro-oncologists and palliative specialists on earlier versus later referral as above, potential barriers to referral were explored. Three themes were identified, each with potential for amelioration. These barriers and strategies for mitigation, with supporting quotes, are in Table 3.

Table 3.

Barriers to Specialty Palliative Care Referral and Potential Strategies to Overcome Them

Barrier to Specialty Palliative Care Referral	Potential Mitigators and Relevant Quotes
Palliative care referrals impose financial and logistical burdens on functionally impaired patients.	• Telehealth: “The clinic ultimately ended up being primarily virtual…which worked quite well, and we would stagger visits [between palliative care and neuro-oncology].”—SPC #9, internal medicine, embedded clinic experience • Embedded palliative care: “[O]ne of the barriers—which then we were able to help with by having an embedded model—was that our palliative care clinic was on a completely different campus...and patients just don’t want to go try to find a whole ‘nother place.”—SPC #9 internal medicine, embedded clinic experience • Joint visits: “The best conversations we’ve had are when we’ve been able to get the oncologist or the neuro-oncologist on the phone.”—SPC #8, internal medicine, no embedded clinic
Stigma equating palliative care with end-of-life care is heightened in HGG due to death anxiety and the reality that discussing goals of care is a priority for this population.	• Earlier referral to facilitate rapport-building while patients are doing well: “I’m a believer that palliative care, particularly in the outpatient setting, should just be concurrent with cancer directed therapy…it, I think, helps to know us, know who we are. And then as things evolve, we develop a relationship that makes it then easier if things do decline to have those kinds of conversations and re-address goals of care.”—SPC #5, internal medicine, no embedded experience • Frame palliative care as part of the team: “There are symptoms that will double because of the disease and because of treatment, and in those stages we will need for the help to be different, with the experts, and I will put palliative care as part of it. So, I wouldn’t just put palliative care separately. So, for example, if you are doubling side effects I may ask in a gastroenterologist. If you have several symptoms where you have coping issues, we will add in palliative care.”—NO #4, internal medicine, no SPC fellowship • Establish appropriate referral criteria: “…if there’s no acute symptoms, it gets awkward to have mandated pathways…I’d like more of a negotiated discussion with the oncologists about, ‘Here are the things that we can be helpful for. Think about us.’”—SPC #10, internal medicine, no embedded experience • SPC skill in navigating sensitive topics: “I don’t rush to [dive into goals and treatment preferences]. I like to build a therapeutic rapport with loved ones first. Obviously talking about death brings up all kinds of emotions and fears for people so it’s really individually tailored depending on where they’re at and what they’re open to.”—SPC #3, internal medicine, embedded clinic experience
Opportunities for symptom management are limited due to neuropsychiatric symptoms in HGG falling outside the usual palliative care canon.	• Palliative specialists have time to address complex and numerous symptoms: “If they have a constellation of symptoms that are due to terminal illness plus everything else…they need assistance with this, that, and the other, then that’s where I think palliative care can provide our patients the greatest benefit.”—NO #10, neurology, no SPC fellowship • Palliative specialists may uncover symptoms that are not revealed to neuro-oncologists: “[I]n terms of just symptom management…things we were overlooking because we just weren’t asking about it…the palliative care team was really, really good at making sure that the patient was eating well, was on appropriate medications, and appropriate bowel regimen…it ended up significantly improving his quality of life.”—NO #1, neurology, no SPC fellowship • Palliative specialists may offer support and anticipatory guidance: “I focus mostly on ensuring that the patient or the family has a good understanding of their prognosis based on what I have gleaned from the rest of the care team, including my own assessment; not just in terms of life expectancy; actually, usually, that’s not what we discuss. It’s mostly about functional status, and what is your life going to look like when this stuff settles.”—SPC #7, internal medicine, no embedded experience

NO = neuro-oncology; SPC = specialty palliative care.

^*Quotes are attributed by participant number (randomly assigned during the analysis phase to maintain privacy) and residency training. Neuro-oncologists are further characterized by whether or not they completed specialty palliative care fellowship, and palliative specialists are characterized by whether or not they have embedded clinic experience.

Firstly, it can be financially and logistically burdensome to attend medical appointments not essential to cancer-directed treatment, especially for neurologically impaired HGG patients. Strategies to mitigate this barrier include the use of telehealth, home-based palliative care, and palliative medicine specialists embedded in neuro-oncology clinics. Joint visits with both neuro-oncology and palliative specialists were also identified as aspirational, but typically not achievable outside of the inpatient setting.

Secondly, stigma associating palliative care with the end of life may seem to be validated given the poor prognosis and relatively limited role of palliative medicine specialists in symptom management. Earlier referral paradoxically addresses this barrier, as it allows time for rapport-building and addressing other issues such as coping. Additionally, palliative medicine specialists are sensitive to death anxiety and skilled at navigating the stigma. When referrals do occur, it is the responsibility of palliative medicine teams to coordinate with neuro-oncologists for accurate prognostication and permission to disclose it. As noted above, it may at times be beneficial for patients to discuss advance care planning with someone outside of their “medical home” in neuro-oncology.

The final identified barrier was that HGG patients’ symptoms require neurologic expertise most palliative specialists do not have. However, this is not universally true. Some patients may have a high burden of non-neurologic symptoms, some of which may not be discovered during neuro-oncology visits. Palliative specialists with experience in embedded clinics emphasized that their roles did frequently include symptom management. Even when options are limited, strategies for mitigating this barrier include coping support and anticipatory guidance.

Discussion

The findings of this study support the hypothesis that, in comparison to other aggressive cancers, optimal palliative care delivery in HGG requires a greater balance of primary palliative care by neuro-oncologists. In addition to having the neurologic expertise needed to manage common symptoms and prognosticate in HGG—skills that are typically not shared by palliative medicine specialists—neuro-oncologists have a unique relationship with their patients as their main partner in treating an aggressive, incurable disease. The importance of primary palliative care also extends to the neuro-oncology team and ancillary services, which may address other palliative care domains (eg, psychosocial and spiritual/existential aspects of care) without necessitating an additional referral. Yet there are potential drawbacks to relying solely on primary palliative care. Neuro-oncologists may not have the time or training to address the totality of palliative care needs in these complex patients.²² Thus, interventions are likely needed to enhance primary palliative care delivery in HGG. Existing examples in neuro-oncology include nurse-led advance care planning discussions between chemoradiation and adjuvant therapy²³; care coordinator-led delivery of regular screening, information, communication and coordination, and care partner engagement at key transition points in the disease trajectory²⁴; and multidisciplinary (neuro-oncology, nursing, psychology, rehabilitation, and social work) home care throughout the disease course and in lieu of hospice at the end of life.²⁵ These interventions vary in scope of palliative care domains addressed and in resources needed for implementation, but all leverage the existing, central role of neuro-oncology practices in providing care to patients with brain tumors.

Yet, the close relationship that develops between neuro-oncologists, patients, and/or caregivers may paradoxically be an impediment to addressing goals of care. This suggests parallels to the concepts of “clinical momentum” in the intensive care unit or “surgical buy-in”: in both cases, an initial agreement to pursue life-prolonging treatment implicitly carries forward to all future decisions and may lead to avoidance of conversations about end-of-life care and, ultimately, to nonbeneficial or unwanted treatment at the end of life for some patients.^26,27 Given the low rates of advance care planning documentation and high healthcare utilization at the end of life among patients with HGG,^13,28,29 this warrants further investigation and may support earlier specialty palliative care referral to facilitate advance care planning with a different clinician and/or team. However, an additional key finding of this study is that the role of specialty palliative care referral in HGG may not be well defined.

For patients who have a high, time-consuming burden of palliative care need (eg, numerous symptoms, complex psychosocial background, or difficulty with decision-making), the decision to refer may be relatively straightforward based on anticipation of tangible benefit (ie, relief of symptoms, support with coping, or greater clarity of goals and values). Given that prior studies have documented high levels of unmet need throughout the HGG illness trajectory, it is likely that more patients fall into this category than is reflected by low referral rates.³⁰ Additionally, participants identified potential roles for palliative medicine specialists throughout the illness trajectory, at times addressing needs that are less tangible or obvious. The question of “what’s in the palliative care syringe” may thus be more nuanced and complex in this population, but this need not be interpreted as a lack of benefit.³¹

To increase specialty palliative care referral rates, consideration of disease-specific barriers to referral is also needed. When to refer relative to diagnosis and other clinical milestones was the issue of greatest contrast between neuro-oncologists’ and palliative specialists’ perspectives, highlighting the need for communication and collaboration between these groups to optimize care delivery. Additionally, participants identified strategies for mitigating the burden of additional appointments for functionally impaired patients, potential for worsening death anxiety, and need for neurologic expertise. Notably, many of these strategies require systems-level change (such as embedded clinics or home-based palliative care to address appointment burden) or greater collaboration between neuro-oncologists and palliative medicine specialists (establishing mutually agreed-upon referral criteria or reframing palliative specialists as “part of the team”).³² Thus, while the barriers to specialty palliative care referral are likely unique to HGG, the solutions may depend on the circumstances of each institution.

This study has several limitations. While the study sample included a diverse group of neuro-oncologists and palliative medicine specialists, they were primarily recruited from academic centers in urban areas. Given that the findings suggest structure of care may be a significant factor in determining the role of palliative medicine specialists, themes may be different in nonacademic and/or rural care settings. It also does not include other key stakeholders: further studies are needed to explore perspectives of HGG patients and caregivers, as well as other clinicians involved in their care (eg, radiation-oncologists, neurosurgeons, nurses, social workers, chaplains, etc.). Allied health providers play key roles in delivering both primary and specialty palliative care,² including in HGG as highlighted by the participants in this study. Understanding these other practitioners’ experiences—which may differ meaningfully from those of physicians—is likely necessary to successfully implement comprehensive palliative care in HGG. Additional next steps in research may include quantitative studies to analyze Patient Reported Outcomes (PROs) and other measurable elements of primary and specialty palliative care delivery in HGG, and ultimately clinical trials of primary and specialty palliative care interventions.^24,33–35

Conclusion

Optimizing the integration of palliative care into the landscape of neuro-oncology can improve the quality of care provided to patients and families affected by this devastating neurologic disease, as it does in nonneurologic cancer. However, the unique clinical context of HGG adds a layer of complexity to palliative care delivery, recognized by both neuro-oncologists and palliative medicine specialists. This study represents an important first step in better understanding the interplay of primary and specialty palliative care in HGG and identifying paths to better care models.

Funding

This work was supported by the National Institute on Aging [AG074064-01]. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. Additional support was provided by the Claude D. Pepper Older Americans Independence Center at the Icahn School of Medicine at Mount Sinai [5P30AG028741]; the Mount Sinai Institute for Health Equity Research; the National Palliative Care Research Center Kornfeld Scholars Program; and the Samuels Foundation.

Conflict of interest statement

None declared.