A Pilot Study of a Collaborative Palliative and Oncology Care Intervention for Patients with Head and Neck Cancer

Abstract

Background:

Palliative care improves symptoms and coping in patients with advanced cancers, but has not been evaluated for patients with curable solid malignancies. Because of the tremendous symptom burden and high rates of psychological distress in head and neck cancer (HNC), we evaluated feasibility and acceptability of a palliative care intervention in patients with HNC receiving curative-intent chemoradiation therapy (CRT).

Methods:

This was a prospective single-arm study in HNC patients receiving CRT at a single center in the United States. The intervention entailed weekly palliative care visits integrated with oncology care with a focus on symptoms and coping. The primary outcome was feasibility, defined as a >50% enrollment rate with >70% of patients attending at least half of the visits. To assess acceptability, we collected satisfaction ratings post-intervention. We also explored symptom burden, mood, and quality of life (QOL).

Results:

We enrolled 91% (20/22) of eligible patients. Patients attended 133 of 138 palliative care visits (96%); all 20 attended >85% of visits. Eighteen of 19 (95%) found the intervention “very helpful” and would “definitely recommend” it. QOL and symptom burden worsened from baseline to week 5, but subsequently improved at one-month post-CRT. Overall, patients valued the one-on-one format of the intervention and receipt of additional care.

Conclusions:

Our palliative care intervention during highly morbid CRT was feasible and acceptable with high enrollment, excellent intervention compliance, and high patient satisfaction. Future randomized studies will further explore the impact on patient-reported outcomes and health care utilization.

Introduction

Patients receiving curative chemoradiation therapy (CRT) for head and neck cancer (HNC) undergo some of the most intensive and morbid regimens employed in the outpatient setting.^1,2 This is particularly challenging in the new era of human-papilloma virus (HPV)-related HNC, which affects a healthier, younger population.³ CRT involves daily radiation for up to seven weeks, paired with weekly chemotherapy and intravenous fluids, with toxicities intensifying over the treatment course. Patients develop significant symptoms, including pain, difficulty swallowing, nausea, and dehydration; more than half require a feeding tube for nutrition.^4,5 These severe symptoms lead to a marked decline in quality of life (QOL) and physical function, and trigger high rates of anxiety and depression in up to 40%–50% of patients, higher than any other cancer population.^6–11

In addition, as many as 44% of patients require hospitalization and up to 53% require dose reductions.¹² Improved symptom assessment and management can decrease rates of hospitalization and improve receipt of chemotherapy^13,14; however, there are few studies of supportive care interventions carried out during CRT when patients may need the enhanced care most.¹⁵ In addition, there is evidence that adherence to supportive care interventions may be lower during CRT due to the intensity of treatment.¹⁶ Thus, effective and timely interventions that are well integrated into the schedule of CRT are needed to improve symptom burden, mood, and QOL.

Palliative care clinicians specialize in pain, complex symptom management, and psychosocial support. Recent studies have demonstrated clinically significant benefits of integrating early palliative care with oncology care for patients with advanced cancer, including improvements in QOL, symptom burden, mood, coping, and survival.^17–22 Despite these advantages, palliative care has been underutilized in patients with HNC, with one study reporting only 5% of patients being seen by a palliative care team.^23,24

Since these patients are receiving curative-intent therapy, they have not been included in early-integrated palliative care trials.^17,18,21,22 In addition, many palliative care interventions to date are consultative and involve monthly visits; however, the cumulative symptom burden in CRT likely necessitates more frequent touch points. Finally, a consultative model with separate appointments may add more burden to an already intense daily schedule. Because of these barriers, innovative models of delivering palliative care interventions are needed.

We conducted a pilot trial of a collaborative palliative care intervention for HNC patients receiving CRT. The intervention entailed a visit with a palliative care team during the first week of CRT followed by weekly patient-centered, nurse-led palliative care visits integrated into the patients' treatment schedules, which focused on coping, mood, and symptom management. The aims were to assess the feasibility and acceptability of the intervention as well as explore symptom burden, mood, and QOL.

Methods

Study design and patient participants

We conducted a prospective pilot study of an outpatient collaborative palliative and oncology care intervention integrated with standard CRT in patients with HNC. Potential participants were identified and recruited from the Multidisciplinary Head and Neck Clinic at Fox Chase Cancer Center (FCCC). Inclusion criteria included (1) patients' age ≥18 years, with a new diagnosis of HNC for which they were to receive CRT, (2) all oncology care performed on the main FCCC campus, (3) able to speak and read in English. Exclusion criteria included (1) prior history of HNC, (2) presence of uncontrolled psychiatric disorders or other comorbid disease, which the clinician believed prohibited consent or participation, and (3) patients enrolled on other intervention trials.

All new HNC patients were reviewed by the study principal investigator (PI) and research assistant to assess eligibility. Written informed consent was obtained at a clinic visit before initiating CRT. All procedures, protocols, and consent forms were approved by the FCCC Institutional Review Board and the study was registered on clinicaltrials.gov (NCT03760471).

Intervention content

All patients received usual oncology care during CRT in combination with the palliative care intervention (Fig. 1). We designed an evidence-based symptom management and coping intervention targeting the most prevalent symptoms that develop during CRT based on prior literature and our own studies.^7,8,25,26 The intervention content was guided by prior research in supportive care interventions with input from medical and radiation oncologists and palliative care clinicians. The intervention visits focused on coping in addition to the following prevalent symptoms: (1) pain and mucositis, (2) nausea, (3) constipation, (4) fatigue, (5) sleep disturbances, (6) xerostomia, (7) thick mucus, and (8) depression.

FIG. 1.

Study schema. CRT, chemoradiation therapy.

During the first week of CRT, patient participants met with a palliative care collaborative team as the initial intervention visit. Each collaborative team consisted of a primary palliative care registered nurse (RN), a supervising palliative care medical doctor (MD) or nurse practitioner (NP), and a social worker (SW). After the first visit, the patients met weekly with the RN, and the MD, NP, and SW were available as needed. During the recovery period, the RN also called each patient weekly for four weeks. All visits were coordinated with the patient's radiation schedule and the palliative care RN conducted each visit in the radiation oncology department or while the patient was receiving hydration. During each visit, the patient and RN created a symptom management plan. This was primarily an outpatient intervention, but if a patient was hospitalized, the palliative care team also followed patients during the admission.

Intervention fidelity

Owing to variability in individual palliative care clinician experience, all participating clinicians underwent training and supervision throughout the study to ensure consistent provision of the intervention. During the training, the PI reviewed the CRT course, the intervention guide, and expectations for visits and documentation. The PI conducted monthly meetings with the palliative care clinicians throughout the trial to review issues pertaining to intervention delivery.

To ensure fidelity, the PI and palliative care MDs reviewed documentation of the weekly RN visits to confirm coverage of the intervention components. In addition, the palliative care clinician completed a weekly visit summary of the appointment content in Research Electronic Data Capture (REDCap), which was reviewed by the PI for intervention adherence and as a comparison between providers.²¹ The visit summary included time spent, topics discussed, and the symptom management plan. The PI provided feedback as needed to individual clinicians to ensure consistent delivery of the intervention.

Measures

Participants completed assessments at baseline (up to four weeks before the start of CRT), at week 5 during CRT (±2 days), and at one-month post-intervention (±2 weeks). Measures included the following:

• 1.Demographics/Disease characteristics: At baseline, participants self-reported age, gender, race/ethnicity, marital status, work status, and education. Clinical, disease, and treatment characteristics (chemotherapy and radiation received, feeding tube placement, and hospitalizations) were obtained from the electronic medical record (EMR).
• 2.Symptom burden: We used the MD Anderson Symptom Inventory-Head and Neck (MDASI-HN) to measure symptom burden. This 28-item questionnaire, well validated and widely used in HNC,²⁷ consists of three subscales: 13 core MDASI symptoms common to all cancers, 9 HNC-specific symptoms, and 6 items that assess how symptoms interfere with daily activities. The items are rated on a 0–10 scale to indicate the symptom severity with 0 indicating “not present” and 10 indicating “as bad as you can imagine.” The interference items are measured 0–10 as well, with 0 indicating “did not interfere” and 10 indicating “interfered completely.” Higher total scores indicate a greater symptom burden or interference. In addition to baseline, week 5 of CRT, and one-month post-CRT, MDASI-HNs were also obtained weekly during CRT and provided to the palliative care provider before the study visit.
• 3.QOL: We used the Functional Assessment of Cancer Therapy-Head and Neck (FACT-HN) to assess QOL. This is a validated 39-item scale that measures four dimensions of QOL (physical, functional, emotional, and social well-being) with 12 head and neck-specific items.²⁸ The items are rated on a 0–4 scale with 0 indicating “not at all” and 4 indicating “very much.” Patients rated the item based on how true the statement was over the last week. Higher scores indicate higher QOL.
• 4.Mood: We used the Hospital Anxiety and Depression Scale (HADS) to assess symptoms of depression and anxiety. This 14-item measure has been used extensively in patients with cancer.²⁹ It is divided into subscales assessing symptoms in the last week of anxiety (HADS-A) and depression (HADS-D), with scores ranging from 0 to 21. Higher scores indicate more distress, with scores greater than seven indicating clinically significant symptoms.
• 5.Acceptability: Intervention acceptability was assessed at one-month post-CRT using two closed-ended items (whether participants found palliative care helpful and whether they would recommend them to others). Response options ranged from 1 indicating “not helpful” to 4 indicating “very helpful,” and 1 indicating “would not recommend” to 4 indicating “definitely recommend.” In addition, to inform future refinements of the intervention, we included five open-ended questions that asked participants their perceptions of the intervention: what they liked and disliked of visits, comments on visit frequency, concerns that were not addressed, and to offer any other feedback regarding the chemoradiation experience.

Data collection

All participant-reported data as well as data from the EMR were entered electronically using REDCap, which is a Health Insurance Portability and Accountability Act (HIPAA) compliant web-based tool commonly used for data capture and management. All participants completed study assessments on paper forms in clinic. If not completed in clinic, participants were mailed surveys to complete at home. Surveys that were not returned were coded as missing data.

Study outcomes

Primary endpoint

The composite primary endpoint for this study was feasibility, and was defined as a >50% enrollment rate and >70% of the participants attending at least half of the scheduled palliative care visits.³⁰

Data analysis

We tested whether the feasible enrollment rate was 50% and whether 70% attended at least half of scheduled visits using two separate one sample binomial exact tests, powering the study to confirm the visit rate among those enrolled. With 20 enrolled patients, we had 89% power to distinguish a discouraging (null hypothesis) rate of 35% with a promising rate of 70%. This assumed the use of a one sample binomial exact test, 89% power, and <2.5% type 1 error rate (one-sided). We applied a Bonferroni correction to the 5% type 1 error rate since we had two primary endpoints (5%/2 = 2.5% one-sided).

For psychosocial measures, we report the mean and standard deviation (SD) at each time point. For visualization and comparisons between time points, we also estimated the means, 95% confidence intervals, and differences (i.e., Betas) using a multiple linear regression with time entered as a categorical variable (i.e., indicator [1/0] variable for time points). To account for the repeated measures nature of the data, we developed the models using Generalized Estimating Equations with an unstructured working correlation matrix.

For secondary analyses, we set the criteria for statistical significance as p < 0.05. Analyses were conducted using STATA v15 (STATACorp, College Station, TX). Participant responses to the open-ended items were examined and summarized with respect to the nature of the feedback (e.g., pertaining to the format of the intervention and the content) to inform potential refinements for a future study.

Results

Patient characteristics

We screened a total of 36 patients from January 2019 to September 2019 for study eligibility and 22 were determined to be eligible and approached (Fig. 2). Twenty enrolled for a 91% enrollment rate (p < 0.0001). Patients were predominantly male (90%), white (90%), employed (90%), and married (85%) with a median age of 59 (39–74) (Table 1). The most common disease subsite was oropharynx (12/20, 60%), all of whom had HPV-positive disease. Sixteen of the 20 patients (80%) were being treated with definitive chemoradiation without prior surgery, whereas four were being treated with adjuvant chemoradiation after surgery. All 20 patients completed baseline and week 5 measures, and 19/20 (95%) completed the assessment at one month after the end of CRT.

FIG. 2.

Consort diagram. HNSCC. CRT, chemoradiation therapy; HNSCC, head and neck squamous cell carcinoma.

Table 1.

Patient Characteristics

Characteristic	Total: n (%) or median (range)
Male	18/20 (90)
Age, median (range)	59 (39–74)
Tumor subsite
Oral cavity	3/20 (15)
Oropharynx	12/20 (60)
Hypopharynx	2/20 (10)
Larynx	3/20 (15)
Clinical stage
I/II	9/20 (45)
III/IV	11/20 (55)
ECOG performance status
0	19 (95)
1	1 (5)
HPV positive, oropharynx	12/12 (100)
Smoking history
Never	13/20 (65)
1–9 years	0 (0)
≥10 pack years	7/20 (35)
Treatment plan
Definitive chemoradiation	16/20 (80)
Adjuvant chemoradiation	4/20 (20)
Chemotherapy type
High-dose cisplatin	7 (35)
Weekly cisplatin	10 (50)
Carboplatin and paclitaxel	2 (10)
Carboplatin	1 (5)
Feeding tube needed	7/20 (35)
Patient education
High school or below	9/20 (45)
Part college or higher	11/20 (55)
Race
African American or Black	2/20 (10)
White	18/20 (90)
Relationship status
Married	17/20 (85)
Employment status
Part or full time work	18/20 (90)

ECOG, Eastern Oncology Cooperative Group; HPV, human-papilloma virus.

The majority of patients received cisplatin as their radiation sensitizer (17/20, 85%). Only one patient had a prolonged radiation course due to deferred fractions from toxicity. For all patients, 12/20 (60%) had a dose reduction or missed dose of chemotherapy due to toxicities. Seven of 20 patients (35%) required a feeding tube during treatment. Five patients (25%) required hospitalization during treatment.

Palliative care visits and feasibility

Patients attended 133 out of 138 possible palliative care visits (96%), and all 20 attended >85% of appointments, meeting the feasibility primary endpoint (p < 0.0001). For the five visits that were unable to be completed, four were due to the patient needing to leave before palliative could see them and one was due to a hospitalization. The initial palliative care visit lasted an average of 43 minutes (SD 14.4), and follow-up appointments lasted an average of 35 minutes (SD 16.0). A caregiver was present in 9/20 (45%) of initial palliative care visits and in 39/113 (35%) of follow-up palliative care visits.

Table 2 depicts topics and symptoms addressed during palliative care visits. Most commonly, palliative care visits centered on rapport-building, symptom management, illness understanding, and coping. The symptoms addressed in >40% of patients were pain and/or mucositis, nausea/vomiting, constipation, fatigue, xerostomia, and/or thick mucus.

Table 2.

Palliative Care Visit Content

Visit content and symptoms addressed	Initial palliative care visit (n = 20), n (%)	Follow-up palliative care visits (n = 113), n (%)
Visit content
Rapport building	20 (100)	78 (69)
Symptom management	13 (65)	105 (93)
Illness understanding	12 (60)	39 (35)
Patients' goals/values	8 (40)	17 (15)
Coping	12 (60)	46 (41)
Advance care planning	0 (0)	0 (0)
Disposition (ER, etc.)	0 (0)	1 (1)
Symptoms addressed
None	2 (10)	4 (4)
Pain and/or mucositis	11 (55)	82 (73)
Nausea/vomiting	9 (45)	52 (46)
Constipation	6 (30)	50 (44)
Fatigue	2 (10)	50 (44)
Sleep disturbances	4 (20)	30 (27)
Xerostomia	4 (20)	51 (45)
Thick mucus	1 (5)	44 (39)
Depression	2 (10)	6 (5)
Other	1 (5)	20 (18)

ER, emergency room.

Acceptability

The response rate at the one-month assessment post-CRT was 95% (19/20). One participant did not return the post-CRT assessment. Among the 19 respondents, 18/19 (95%) patients reported they found the intervention to be “very helpful” and 18/19 (95%) would “definitely recommend it to others.”

In the open-ended questions, patients commented on liking the format of the palliative care visits through the “one-on-one nature,” “having someone to discuss symptoms/reactions in addition to visits with doctors,” and how “convenient and accommodating all visits were.” They also appreciated the content of the palliative care visits, with comments on the “good advice” received, the “insight” into the patient's illness and experience, and the general “extra layer of support” provided. Patients did not report disliking any aspect of the palliative care visits, and all reported that the weekly visit frequency was appropriate. Other general feedback from participants about how to improve the experience of patients receiving CRT included suggestions on natural remedies such as aloe vera or honey as well as improving the logistical nature of the center with valet parking.

Patient-reported outcomes

Figure 3 and Table 3 depict patient-reported QOL, symptom burden, and mood at baseline (T0), week 5 (T1), and one-month post-CRT (T2). As expected during a course of chemoradiation, QOL (FACT-HN) worsened from baseline to week 5 (β −20.19, p ≤ 0.0001) and then started to improve at one-month post-CRT (β 6.82, p = 0.0226). Symptom burden (MDASI-HN) increased from baseline to week 5 (β 1.52, p = 0.0001) and then started to improve at one-month post-CRT (β −1.07, p ≤ 0.0001).

FIG. 3.

Patient-reported mean outcomes at baseline, week 5, and one month after CRT. (A) Quality of life, (B) symptom burden, (C) depression, and (D) anxiety. FACT, Functional Assessment of Cancer Therapy; HADS, Hospital Anxiety and Depression Scale; MDASI, MD Anderson Symptom Inventory.

Table 3.

Patient-Reported Outcomes at Baseline (T0), Week 5 (T1), and One Month after Chemoradiation Therapy (T2) and Differences (Betas) between Time Points

	T0 mean (SD), n = 20	T1 mean (SD), n = 20	T2 mean (SD), n = 19	T1–T0 β	p	T2–T0 β	p	T2–T1 β	p
HADS-depression	2.54 (2.88)	5.85 (3.47)	4.89 (4.03)	3.31	0.0000	2.20	0.0075	−1.11	0.0522
HADS-anxiety	4.05 (3.85)	3.81 (3.11)	3.68 (3.38)	−0.24	0.6047	−0.43	0.4898	−0.19	0.7485
FACT-HN	117.88 (20.48)	97.69 (20.27)	104.21 (23.72)	−20.19	0.0000	−13.37	0.0040	6.82	0.0226
MDASI-HN	1.47 (1.34)	2.99 (1.62)	2.01 (1.50)	1.52	0.0001	0.45	0.2074	−1.07	0.0000

FACT-HN, Functional Assessment of Cancer Therapy-Head and Neck; HADS, Hospital Anxiety and Depression Scale; MDASI-HN, MD Anderson Symptom Inventory-Head and Neck; SD, standard deviation.

Anxiety remained similar across time points (T1–T0 β −0.24, p = 0.604; T2–T1 β −0.19, p = 0.7485), but depression increased from baseline to week 5 (β 3.31, p = 0.0000) and remained higher at one-month post-CRT (β 2.20, p = 0.0075). At baseline, 1/20 (5%) of patients had clinically significant depressive symptoms as measured by the HADS, 6/20 at week 5 (30%), and 4/19 (21%) at one-month post-CRT.

Discussion

In this study of patients receiving curative-intent CRT for HNC, integrating a patient-centered, nurse-led palliative care intervention weekly into treatment was both feasible and acceptable, meeting the protocol-specified criteria for study success. The high enrollment, visit adherence, and acceptability demonstrate a clear perceived need for enhanced supportive care during CRT and supports our patient-centered model.

To our knowledge, this is the first study of a palliative care-specific intervention integrated into outpatient curative-intent treatment for HNC. Previous interventions for HNC patients have primarily included educational^25,31–33 or psychological interventions^34–38 to improve symptoms, and often occurred after treatment had ended.^36,39 Despite palliative care's expertise in complex symptom management and evidence that palliative care interventions improve patient QOL, mood, symptom burden, and coping in patients with advanced cancer,^17,18,21,22 evidence is needed to support palliative care integration into the curative solid tumor setting to overcome barriers to integration and convince payers and health care systems to invest in these models.

Palliative care has likely not been previously integrated into CRT for HNC due to a combination of factors: misconceptions by both patients and clinicians equating palliative care to hospice; a lack of access to comprehensive palliative care services in the outpatient setting; underappreciation of the substantial and ameliorable symptom burden in HNC; and reluctance to increase patient obligations during rigorous and time-consuming therapy. Thus, the successful feasibility of our intervention provides proof of concept for palliative care integration both in the curative outpatient setting and in the HNC population.

Notably, our study used a nurse-driven model with weekly instead of monthly visits for the palliative care intervention. Prior models of monthly palliative care visits are likely too infrequent to impact the extreme symptom burden of CRT. Moreover, prior palliative care interventions have often used physicians and/or NPs to carry out the intervention,^18,21,22 but there is a shortage of palliative care trained clinicians and thus weekly visits would not be feasible.⁴⁰

Nurses, however, have successfully carried out supportive care interventions with HNC patients, in particular when coordinated with treatment visits or at patients' homes, and they have been involved in the delivery of palliative care interventions in advanced cancer.^15,17 Thus, using a palliative care RN as the cornerstone of the intervention, with access as needed to a larger multidisciplinary palliative care team, carries the potential to increase comprehensive symptom management in a sustainable, accessible, and cost-effective manner, which could lead to easier dissemination.

As expected during an intense course of CRT, QOL and symptom burden worsened during treatment at week 5 and then started to improve at one month after CRT ended in our patient cohort. Interestingly, the declines in QOL observed in this study appear to be somewhat less pronounced than those that have been historically reported in patients undergoing chemoradiation.⁴¹ In addition, only 30% of patients had clinically significant depressive symptoms at the end of treatment, which is also lower than other studies that have reported rates of depressive symptoms in up to 40%–50% of patients.^9,11

Although investigators should limit cross-study comparison, the findings suggest that this palliative care intervention may be attenuating the physical and psychological symptom burden of patients with HNC receiving CRT. In addition, only 25% of patients were hospitalized during this study, which is lower than previously reported rates of 44%.¹² Thus, future randomized studies exploring whether the intervention does attenuate the physical and psychological symptom burden and reduce health care utilization in this population will be important for patients and health care systems alike.

Important limitations warrant attention. This study was undertaken at a single comprehensive cancer center with a dedicated head and neck multidisciplinary team and robust outpatient palliative care services; even large general hospitals may not provide the same resources. The study population was predominantly male, white, and married, potentially reducing the generalizability of the findings to more diverse populations. In addition, acceptability was measured using several open- and closed-ended questions; however, an in-person qualitative interview may have offered additional insight on intervention acceptability.

All visits to oncology physicians, chemotherapy, radiation, and palliative care occurred in the same building, and the palliative care nurses traveled to the patient for visits, providing true patient-centered care. Hence, streamlined coordination of care and the layout of our center may have helped to achieve intervention compliance and feasibility. Other centers with greater physical separation of points of care, or nurses with less flexibility, may have more difficulty carrying out such an intervention.

Similar to other palliative care studies,^18,26 we tracked intervention fidelity using the palliative care clinician's report of topics covered during their visits and the PI routinely reviewed this documentation. However, we acknowledge that self-report is a potential source of bias and that audio recording to examine fidelity with the PI reviewing randomly selected recordings would have been a more objective measure and will be included in future iterations.

Finally, this is a single-arm study without a comparison arm. A small single-arm design can have problems with internal validity as there are issues with selection bias, history, and maturation threat, which limit our ability to determine whether the intervention impacted patient-reported outcomes of QOL, mood, and symptom burden. However, before implementing a two-arm study, it was critical to demonstrate that we could feasibly carry out a weekly palliative intervention led by nurses and coordinated in tandem with clinical visits, which we accomplished in this initial single-arm study. These findings will be used to support the design of a larger randomized study to rigorously assess the impact of this intervention on patient outcomes.

Conclusion

In conclusion, our pilot palliative care intervention during CRT for patients with HNC proved both feasible and acceptable with a high enrollment rate, excellent intervention compliance, and high patient satisfaction. We successfully integrated palliative care systematically into the care of patients with HNC and into outpatient curative treatment, using a palliative care nurse as the cornerstone of the intervention. The success of this pilot study establishes feasibility and acceptability for better supportive care during highly morbid curative-intent treatment, setting the stage for larger randomized studies to further evaluate the impact of palliative care interventions on patient-reported outcomes and health care utilization.

Funding Information

This project is supported by the Palliative Care Research Cooperative Group (PCRC) funded by National Institute of Nursing Research U2CNR014637. The core grant at Fox Chase Cancer Center NIH/PCI P30CA006927 also supported this study.

Author Disclosure Statement

J.R.B. has received payments for consulting for Bayer, Pfizer, Kura, and Astra Zeneca, as well as research funding from BMS to her institution, none of which were related to this research. M.E.C. reports receiving payment for a book on palliative care. J.S.T. reports research funding from Pfizer to her institution.

Prior abstract presentations: Supportive Care in Oncology Symposium, San Francisco, CA, October 2019, Abstract 73; Head and Neck Symposium, Scottsdale, AZ, February 2020, Abstract 372.