Immune Checkpoint Inhibitor Use Near the End of Life Is Associated With Poor Performance Status,Lower Hospice Enrollment,and Dying in the Hospital

Abstract

Background:

Immune checkpoint inhibitors have changed the landscape of cancer care by increasing progression-free and overall survival in some patients with cancer. We evaluated use and variables contributing to immune checkpoint inhibitor treatment near the end of life.

Methods:

We studied 157 patients who received immune checkpoint inhibitors and died between January 2015 and December 2018. All patients had a palliative care consult any time between starting an immune checkpoint inhibitor and death. Univariate and multivariate models were used to examine variables related to immune checkpoint inhibitor use near the end of life.

Results:

Among 157 patients studied, 42 (27%) received a dose of immune checkpoint inhibitor in the last 30 days of life. Those who received treatment in the last 30 days of life had lower hospice enrollment (19 [45%] vs 78 [69%], P = .007) and higher rates of dying in the hospital (23 [56%] vs 33 [29%], P = .002). The percentage of patients with Eastern Cooperative Oncology Group (ECOG) ≥3 at the time of last immune checkpoint inhibitor dose was higher in the group that received immune checkpoint inhibitor treatment in the last 30 days of life (11 [26%] vs 9 [8%], P = .003). Lack of traditional chemotherapy after immune checkpoint inhibitor, ECOG ≥3, and lack of hospice enrollment were independently associated with receiving immune checkpoint inhibitor in the last 30 days of life.

Conclusion:

Immune checkpoint inhibitor use in the last 30 days of life is common and associated with poor performance status, lower hospice enrollment, and dying in the hospital.

Keywords

immune checkpoint inhibitor end of life cancer immunotherapy palliative care oncology

Introduction

Immunotherapy agents referred to as “immune checkpoint inhibitors” have transformed the standard of care for patients with cancer.^1

–5 These agents have been shown to extend overall survival and progression-free survival in patients with lung cancer, melanoma, Hodgkin lymphoma, renal cell carcinoma, urothelial cell carcinoma, Merkel cell carcinoma, head and neck cancer, and more.^3,6
–8 The promise of immune checkpoint inhibitors is their ability to activate the immune system to better recognize and attack cancer cells.⁹ As a result of more precise therapy, patients are able to avoid the systemic side effects of traditional chemotherapy, such as alopecia, nausea, vomiting, fatigue, and myelosuppression.³ Alternatively, immune checkpoint inhibitors have their own set of challenges, such as predicting who will respond, toxicities, and challenging treatment decisions. The advent of new immunotherapy drugs has led to what some are calling “desperation oncology”—referring to the initiation of these drugs as a final effort to stave off death.¹⁰

In the era of traditional chemotherapy, studies revealed that prescribing chemotherapy toward the end of life does not improve quality of life and may result in minimal survival benefits, treatment side effects, increased costs, and higher rates of death in the hospital.^11

–15 As a result of such studies, the use of chemotherapy at end of life has become an indicator for poor quality of cancer care. Subsequent to the “Choosing Wisely” campaign developed by the American Board of Internal Medicine Foundation, the American Society of Clinical Oncology published a top 5 list of opportunities to improve care and reduce costs in oncology. The first item on the list states “don’t use cancer-directed therapy for solid tumor patients with the following characteristics: low performance status (3 or 4), no benefit from prior evidence-based interventions, not eligible for a clinical trial and no strong evidence supporting the clinical value of further anti-cancer treatment.”^16(p1716)

Use of immune checkpoint inhibitors near the end of life has not been described. The nature of cancer response patterns to these agents and prolonged time to response provide new challenges for clinicians regarding symptoms, prognostication, goals of care, and costs.¹⁷ The objective of this retrospective study is to describe the rate and predictors of immune checkpoint inhibitor use near the end of life among patients with advanced-stage cancer.

Methods

We reviewed the electronic medical records of deceased patients who had a palliative care inpatient consult any time after receiving an immune checkpoint inhibitor at our 750-bed academic medical center that includes a comprehensive palliative care program as well as an National Cancer Institute comprehensive cancer center, between January 2015 and December 2018. Patients who were lost to follow up or who transferred care after immune checkpoint inhibitor therapy were excluded. This study was exempt from institutional review board review.

Information gathered for each patient included demographic characteristics (age at death, gender, race, body mass index [BMI]), type of cancer, date of cancer diagnosis, purpose of palliative care consult, immune checkpoint inhibitor medication, date of last dose of immune checkpoint inhibitor, palliative or curative intent of therapy, immune-related adverse events (irAEs), hospice enrollment, date of death, and place of death. Palliative performance status (PPS) scores, Eastern Cooperative Oncology Group (ECOG) scores, and Edmonton Symptom Assessment Scores (ESAS) were recorded from the time of first palliative care inpatient consult after receiving immune checkpoint inhibitor. ECOG score was also recorded at the time of last immune checkpoint inhibitor dose.

Data Analysis

Baseline demographics were reported with descriptive statistics, including frequency, percent, mean, standard deviation (SD), and ranges. Comparisons were made between patients who received immune checkpoint inhibitor less than or equal to 30 days before death and patients who received immune checkpoint inhibitor more than 30 days before death, using χ² or Fisher exact tests, as appropriate for categorical variables and t tests for continuous variables. P value of less than or equal to .05 was used to determine significance. Univariate and multivariate models with logistic regression analysis were used to determine factors contributing to immune checkpoint inhibitor use in the last 30 days of life. All analyses were conducted using SAS software 4.0 (Cary, North Carolina).

Results

We analyzed 157 charts of patients with cancer who had a palliative care inpatient consult after receiving immune checkpoint inhibitor medication and died between January 2015 and December 2018. Table 1 shows the baseline descriptive characteristics of these patients gathered at the time of the inpatient palliative care consult. Sixty-nine (44%) were female, mean age at death was 61 years (SD 13.9; range 28-90 years), 151 (96%) were Caucasian, and 152 (97%) had stage IV cancer. The most common cancer types represented were non-small cel lung cancer 49 (31%), melanoma 39 (25%), urothelial cell carcinoma 12 (8%), and renal cell carcinoma 11 (7%) followed by others 46 (29%). The median interval between last dose of immune checkpoint inhibitor and death was 55 days.

Table 1.

Patient Characteristics (n = 157).

Variable	n (%)^a
Age in years at death, mean (range)	61 (28-90)
Female sex	69 (44)
Race
Caucasian	151 (96)
Other	6 (4)
Cancer stage IV	152 (97)
Type of cancer
NSCLC	49 (31)
Melanoma	39 (25)
Urothelial	12 (8)
RCC	11 (7)
Other	46 (29)

Abbreviations: NSCLC, non-small cell lung cancer; RCC, renal cell carcinoma.

^aColumn percentile unless otherwise designated.

Table 2 shows comparisons between patients who received an immune checkpoint inhibitor in the last 30 days of life and those who received their last dose more than 30 days prior to death. Of the 157 patients in this study, 42 (27%) received an immune checkpoint inhibitor in the last 30 days of life. Patients who received immune checkpoint inhibitor in the last 30 days of life had lower enrollment in hospice (19 [45%] vs 78 [69%], P = .007) and higher rates of dying in the hospital (23 [56%] vs 33 [29%], P = .002). More patients who received immune checkpoint inhibitor in the last 30 days of life had traditional chemotherapy prior to immune checkpoint inhibitor (32 [76%] vs 66 [57%], P = .03), but fewer received chemotherapy after the immune checkpoint inhibitor (2 [5%] vs 47 [41%], P = < .001). There was no statistical difference between the 2 groups for age, BMI, ECOG, PPS, or ESAS at the time of palliative care inpatient consult. There was also no difference in percentage of consults for goals of care, type of cancer, immune checkpoint inhibitor agent, or number of doses. Patients who received immune checkpoint inhibitor in the last 30 days of life had a lower rate of brain metastases (10 [24%] vs 48 [42%], P = .04) and were less likely to require oral steroids for treatment of irAEs. At the time of last dose, mean ECOG score (2.1 vs 1.4, P ≤.001) and rate of patients with ECOG ≥3 (11 [26%] vs 9 [8%], P = .003) were higher in the group that received immune checkpoint inhibitor in the last 30 days of life.

Table 2.

Clinical Characteristics Associated With Patients Receiving Immune Checkpoint Inhibitor Near the End of Life.

Patient Characteristic	Immune Checkpoint Inhibitor in the Last 30 Days of Life, n (%)^a		P Value
Patient Characteristic	Yes, n = 42	No, n = 115	P Value
Female sex	19 (45)	50 (44)	.84
Age at death (mean), years	62	61	.72
BMI (mean)	27	26	.13
ECOG score (mean at time of consult)	3.2	3.0	.36
ECOG score (mean at last dose of immune checkpoint inhibitor)	2.1	1.4	< .001
ECOG score ≥3 (at last dose of immune checkpoint inhibitor)	11 (26)	9 (8)	.003
Palliative performance score (mean at time of consult)	39	45	.06
Edmonton symptom assessment score (mean at time of consult)	33	34	.82
Consult for goals of care	31 (74)	74 (64)	.26
Days from PC consult to death (mean)	31.7	58.8	.09
Type of cancer			.63
NSCLC	11 (26)	38 (33)
Melanoma	10 (24)	29 (25)
Urothelial	5 (12)	7 (6)
RCC	2 (5)	9 (8)
Other	14 (33)	32 (28)
Brain metastases present	10 (24)	48 (42)	.04
Last immune checkpoint inhibitor (medication)			.75
Pembrolizumab	20 (48)	47 (41)
Nivolumab	19 (45)	43 (37)
Other	3 (7)	25 (22)
Doses of immune checkpoint inhibitor (mean)	6.2	7.1	.52
irAE documented
Present	8 (19)	36 (31)	.13
Grade ≥3	2 (5)	16 (14)	.16
Req oral steroid	2 (5)	28 (24)	.01
Req IV steroid	1 (2)	5 (4)	1.00
Treatments
Chemo prior to checkpoint	32 (76)	66 (57)	.03
Chemo After Checkpoint	2 (5)	47 (41)	< .001
Chemo combined with checkpoint	6 (14)	17 (15)	.94
Death in hospital	23 (56)	33 (29)	.002
Hospice enrolled	19 (45)	78 (69)	.007

Abbreviations: BMI, body mass index; ECOG, Eastern Cooperative Oncology Group; irAE, immune-related adverse event; NSCLC, non-small cell lung cancer; RCC, renal cell carcinoma.

^aColumn percentile unless otherwise designated.

Univariate analysis (Table 3) revealed factors associated with immune checkpoint inhibitor use in the last 30 days of life were receiving ≤3 doses of immune checkpoint inhibitor, ECOG ≥3 at time of last dose, lack of traditional chemotherapy after immune checkpoint inhibitor, dying in the hospital, and not being enrolled in hospice. Multivariate regression analysis (Table 4) showed that immune checkpoint inhibitor use in the last 30 days of life was independently associated with lack of traditional chemotherapy after immune checkpoint inhibitor, not being enrolled in hospice, and ECOG ≥3 at the time of the last dose. Specific immune checkpoint inhibitor medications that were given in the last 30 days of life are shown in Table 5.

Table 3.

Univariate Models With Logistic Regression Analysis of Factors Associated With Receiving End-of-Life Immune Checkpoint Inhibitor.

Variable	ICI Treatment Given End of Life, OR (95% CI)
Total # doses 1-3 (reference >3)	2.06 (1.00-4.25)
Chemotherapy after ICI (reference = no)	0.07 (0.02-0.31)
Death in hospital (reference = out of hospital)	3.10 (1.48-6.48)
Hospice enrollment (reference = no)	0.37 (0.18-0.77)
ECOG ≥3 (at time of last immune checkpoint inhibitor dose; reference: ≤2)	4.21 (1.58-11.21)

Abbreviations: CI, confidence interval; ECOG, Eastern Cooperative Oncology Group; ICI, immune checkpoint inhibitor; OR, odds ratio.

Table 4.

Multivariate Models With Logistic Regression Analysis of Factors Associated With Receiving End-of-Life Immune Checkpoint Inhibitor.

Variable	ICI Treatment Given End of Life, OR (95% CI)
Chemotherapy after ICI (reference = no)	0.09 (0.09-0.45)
Hospice enrolled (reference = no)	0.30 (0.12-0.76)
ECOG ≥3 (at time of last immune checkpoint inhibitor dose; reference: ≤2)	4.35 (1.29-14.68)

Abbreviations: CI, confidence interval; ECOG, Eastern Cooperative Oncology Group; ICI, immune checkpoint inhibitor; OR, odds ratio.

Table 5.

Cost of Immune Checkpoint Inhibitor Medications in the Last 30 Days of Life.

Immune Checkpoint Inhibitor Medication	# of Doses Administered	Cost Per Dose ($US)^a	Total Cost ($US)
Pembrolizumab^b	22	11 327	249 194
Nivolumab^c	23	7665	176 295
Atezolizumab^d	2	8881	17 762
Ipilimumab + Nivolumab^e	1	25 140	25 140
Total	48		468 391

^aThese costs based on Redbook 2019 database. Does not include costs for administration, physician services, adverse events, or related hospitalizations.

^bAssuming usual dosing for NSCLC with 200 mg flat dose.

^cAssuming usual dosing for NSCLC with 240 mg flat dose.

^dAssuming usual dosing for NSCLC with 1200 mg flat dose.

^eSingle patient with metastatic neuroendocrine tumor treated with combination ipilimumab 1 mg/kg and nivolumab 3 mg/kg. Calculated assuming 70 kg patient.

Discussion

The rate of immune checkpoint inhibitor use near the end of life was not previously described. Our study demonstrated that more than 1 out of 4 patients receiving immune checkpoint inhibitors had a dose given in the last 30 days of life. End of life immune checkpoint inhibitor therapy was associated with poor performance status, lower hospice enrollment, and dying in the hospital. Previous studies involving chemotherapy use in the last month of life range from 12% to 43% and 14% to 49% for targeted therapy (such as epidermal growth factor receptor (EGFR) and anaplastic lymphoma kinase (ALK) inhibitors used in lung cancer).^{12,18

–27} One study demonstrated that after approval of immune checkpoint inhibitors in urothelial cell carcinoma, patients initiating any systemic therapy in the last 60 days of life increased from 17.4% to 34.8%. They also saw an increase in end of life immune checkpoint inhibitor use among patients with poor performance status ECOG ≥2.²⁸ Given that immune checkpoint inhibitors require several months to result in a tumor response, one might expect that rates of use in the last month would be lower than that of chemotherapy; however, this was not observed in our study. High rates of immune checkpoint inhibitor use near the end of life may be due to perceived tolerability compared to traditional chemotherapy, as evidenced by some of the patients in our study who received treatment at 90 years of age, had performance status ECOG score ≥3, or died as soon as 1 day after a dose. Our study showed that poor performance status is an important prognostic factor when considering immune checkpoint inhibitor therapy, just as previously noted with traditional chemotherapies and by a recent study of immune checkpoint inhibitors in patients with non-small cell lung cancer ≥70 years old.²⁹ Receiving ≤3 doses was also associated with immune checkpoint inhibitor treatment in the last 30 days of life, revealing that patients are often beginning this therapy near the end of their life, which yields little to no clinical benefit.

The decision to provide cancer treatment with an immune checkpoint inhibitor is often complicated, involving clinical factors, performance status, patient preferences, social factors, and physician experience. Unfortunately, the ability to predict which patients and which tumors will respond to immune checkpoint inhibitors remains poor and only a small percentage (20%-40%) of these patients have a positive tumor response.^1,3,30 In our study, treatment near the end of life was associated with lower hospice enrollment, as expected, due to existing capitated hospice reimbursement policies.³¹ There is concern that immune checkpoint inhibitors have little clinical value when given in the last 30 days of life and may negatively affect a patient’s dying experience by delaying what has been described as the “life–death transition process” for patients and families.³² This is a time when patients and families can settle financial affairs, focus on relief of suffering, move through stages of anticipatory grief, and concentrate on meaningful emotional and spiritual closure with loved ones. Higher rates of death in the hospital were also seen in the group receiving immune checkpoint inhibitor in the last 30 days of life, which was consistent with a previous study on chemotherapy near the end of life.¹⁵ The study showed patients receiving palliative chemotherapy near the end of life were more likely to die in the intensive care unit, less likely to die in their preferred place, and often had late hospice referrals.¹⁵

As the field of palliative care and use of immune checkpoint inhibitors both continue to expand, palliative care clinicians will find themselves caring for many of these patients with cancer. Our results demonstrate that when palliative care clinicians encounter patients after receiving immune checkpoint inhibitor therapy, they are extremely ill and have a considerable burden of symptoms. Additionally, these patients commonly have unique, unpredictable toxicities as was seen in more than one-quarter of our study patients that had a documented irAE in their chart. Patients receiving treatment near the end of life had lower incidence of irAEs, potentially because many did not live long enough for irAEs to develop and were often just starting immune checkpoint inhibitor treatment.

Lastly, with these considerable advances in cancer treatment come costs. Immune checkpoint inhibitor treatment given in the last 30 days of life has low clinical benefit and high financial toxicity. In our study, 42 patients received treatment in the last 30 days of life, which lead to increased costs for the medication and administration, subsequent follow-up appointments, hospitalizations, and expensive end of life care. A recent study revealed that in 9.5 million new cancer diagnoses (between 2000 and 2012), 42% of these individuals had depleted their entire net worth only 2 years after diagnosis.³³ Although financial toxicity remains a concern in the US health-care system, use of immune checkpoint inhibitors near the end of life demonstrates an area for improvement and cost savings.

Our study has limitations. Firstly, this study is retrospective and relied heavily on the accuracy of the electronic medical records. Patients at our tertiary care center may have more advanced or refractory cancers than those seen in typical community oncology clinics. Almost all patients were Caucasian, which culturally may impact end-of-life decision-making and treatment. The number of patients included in this study is only 157; thus, further studies are needed to determine more about patterns of immune checkpoint inhibitor use near the end of life. Separating the patients into those that received immune checkpoint inhibitor near the end of life and those that did not using a cutoff of ≤30 days was extremely conservative since time to tumor responses with this therapy is usually several months; the true cost of nonbeneficial doses is likely to be much higher. Table 5 medication costs do not include physician or facility costs, hospitalizations, or adverse events; therefore, the actual cost at the end of life is much higher for these 42 patients who received 48 doses of immune checkpoint inhibitor near the end of life. We recognize a bias when assuming that dying in the hospital is not preferred; however, there are also patients who do not feel comfortable dying at home or who prefer to die in the hospital. Information about place of death, hospice enrollment, PPS, ECOG, and ESAS was more often available but could not be found for some patients. We appreciate that patients who require a palliative care inpatient consult may be generally more ill or have complicated goals of care. Lastly, almost all of our patients had previously treated metastatic cancer, whereas recently immune checkpoint inhibitors have become the first-line treatment for some malignancies like non-small cell lung cancer.

Conclusion

Rates of immune checkpoint inhibitor use near the end of life were similar to that of chemotherapy and targeted therapy. In this study, use of immune checkpoint inhibitors near the end of life was associated with poor performance status, lower hospice enrollment, and dying in the hospital. Further studies are needed to determine more information about the effect of immune checkpoint inhibitors on quality of life, symptoms, performance status, and end-of-life care.

An oncologist’s decision to initiate, continue, or discontinue immune checkpoint inhibitor therapy is complicated and requires consideration of patient preferences, goals of the therapy, disease characteristics, performance status, comorbidities, prognosis, and social support. Immune checkpoint inhibitors have not been studied in and should not be used in patients with ECOG ≥3. These patients may not live long enough to experience a response to immune checkpoint inhibitors and alternatively could focus on spending time with family, completing important final tasks and enrolling in hospice. It is of utmost importance that oncologists and palliative care specialists provide patients clarity through frank communication regarding prognosis and likelihood of response to immune checkpoint inhibitors when it appears that their mortality may be very near.

Footnotes

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Chad Glisch

References

Davis

Panikkar

Checkpoint inhibitors, palliative care, or hospice. Curr Oncol Rep. 2018;20(1):2.

Morabito

. Second-line treatment for advanced NSCLC without actionable mutations: is immunotherapy the ‘panacea’ for all patients? BMC Med. 2018;16(1):24.

Temel

Gainor

Sullivan

Greer

. Keeping expectations in check with immune checkpoint inhibitors. J Clin Oncol. 2018;36(17):1654–1657.

Verma

Sprave

Haque

, et al. A systematic review of the cost and cost-effectiveness studies of immune checkpoint inhibitors. J Immunother Cancer. 2018;6(1):128.

Wang

Gilbert

Kim

. Immunotherapy: who is eligible? Otolaryngol Clin North Am. 2017;50(4):867–874.

Champiat

Ferrara

Massard

, et al. Hyperprogressive disease: recognizing a novel pattern to improve patient management. Nat Rev Clin Oncol. 2018;15(12):748–762.

Jenkins

Barbie

Flaherty

. Mechanisms of resistance to immune checkpoint inhibitors. Br J Cancer. 2018;118(1):9–16.

Thallinger

Füreder

Preusser

, et al. Review of cancer treatment with immune checkpoint inhibitors: current concepts, expectations, limitations and pitfalls. Wien Klin Wochenschr. 2018;130(3-4):85–91.

Berger

. PD-1 pathway and its clinical application: a 20 year journey after discovery of the complete human PD-1 gene. Gene 2018;638:20–25.

10.

Kolata

. ‘Desperation Oncology’: When Patients Are Dying, Some Cancer Doctors Turn to Immunotherapy, in New York Times. New York, NY: The New York Times Company; 2018.

11.

Garrido

Prigerson

Bao

Maciejewski

. Chemotherapy use in the months before death and estimated costs of care in the last week of life. J Pain Symptom Manage. 2016;51(5):875–881.

12.

Näppä

Lindqvist

Rasmussen

Axelsson

. Palliative chemotherapy during the last month of life. Ann Oncol. 2011;22(11):2375–2380.

13.

Prigerson

Bao

Shah

, et al. Chemotherapy use, performance status, and quality of life at the end of life. JAMA Oncol. 2015;1(6):778–784.

14.

Temel

Greer

Muzikansky

, et al. Early palliative care for patients with metastatic non-small-cell lung cancer. N Engl J Med. 2010;363(8):733–742.

15.

Wright

Zhang

Keating

Weeks

Prigerson

. Associations between palliative chemotherapy and adult cancer patients’ end of life care and place of death: prospective cohort study. BMJ. 2014;348:g1219.

16.

Schnipper

Smith

Raghavan

, et al. American society of clinical oncology identifies five key opportunities to improve care and reduce costs: the top five list for oncology. J Clin Oncol. 2012;30(14):1715–1724.

17.

Hansen

Wang

Case

Puzanov

Smith

. Immune checkpoint inhibitor toxicity review for the palliative care clinician. J Pain and Symptom Manage. 2018;56(3):460–472.

18.

Adam

Hug

Bosshard

. Chemotherapy near the end of life: a retrospective single-centre analysis of patients’ charts. BMC Palliat Care. 2014;13:26.

19.

Braga

Miranda

Fonseca

, et al. The aggressiveness of cancer care in the last three months of life: a retrospective single centre analysis. Psychooncology. 2007;16(9):863–868.

20.

Hiramoto

Tamaki

Nagashima

, et al. Prognostic factors in patients who received end-of-life chemotherapy for advanced cancer. Int J Clin Oncol. 2019;24(4):454–459.

21.

Hui

Karuturi

Tanco

, et al. Targeted agent use in patients with cancer at the end of life. J Clin Oncol. 2012;30(34 suppl):215.

22.

Kao

Shafiq

Vardy

Adams

. Use of chemotherapy at end of life in oncology patients. Ann Oncol. 2009;20(9):1555–1559.

23.

Murillo

Jr Koeller

. Chemotherapy given near the end of life by community oncologists for advanced non-small cell lung cancer. Oncologist. 2006;11(10):1095–1099.

24.

Soh

Yuen

Teo

, et al. Targeted therapy at the end of life in advanced cancer patients. J Palliat Med. 2012;15(9):991–997.

25.

Wong

Teo

Lim

Wong

Soo

Chan

. Targeted therapy at the end of life for patients with lung cancer. J Palliat Med. 2010;13(8):945–948.

26.

Zhang

Chen

Liu

Zhao

Cheng

. Palliative chemotherapy near the end of life in oncology patients. Am J Hosp Palliat Care. 2018;35(9):1215–1220.

27.

Zhang

Chen

Liu

Zhao

Cheng

. Use of palliative chemo- and radiotherapy at the end of life in patients with cancer: a retrospective cohort study. Am J Hosp Palliat Care. 2017;34(9):801–805.

28.

Parikh

Galsky

Gyawali

. Trends in checkpoint inhibitor therapy for advanced urothelial cell carcinoma at the end of life: insights from real-world practice. Oncologist. 2019;24(6):e397–e399.

29.

Muchnik

Loh

Strawderman

, et al. Immune checkpoint inhibitors in real-world treatment of older adults with non-small cell lung cancer. J Am Geriatr Soc. 2019;67(5):905–912.

30.

Kato

Goodman

Walavalkar

Barkauskas

Sharabi

Kurzrock

. Hyperprogressors after immunotherapy: analysis of genomic alterations associated with accelerated growth rate. Clin Cancer Res. 2017;23(15):p4242–4250.

31.

Aldridge Carlson

Barry

Cherlin

McCorkle

Bradley

. Hospices’ enrollment policies may contribute to underuse of hospice care in the United States. Health Aff (Millwood). 2012;31(12):2690–2698.

32.

Gyawali

Niraula

. Cancer treatment in the last 6 months of life: when inaction can outperform action. Ecancermedicalscience. 2018;12:826.

33.

Gilligan

Alberts

Roe

Skrepnek

. Death or debt? National estimates of financial toxicity in persons with newly-diagnosed cancer. Am J Med. 2018;131(10):1187–1199e5.