Home-based geriatric oncology care: A feasibility study

62

Background: The Comprehensive Geriatric Assessment (CGA) improves the outcomes of older patients with cancer, and by identifying of geriatric impairments the CGA can guide the development of supportive interventions. However, the implementation of these interventions in a real-world setting can be burdensome for patients, particularly when additional visits are required in an often frail population. Therefore, our team developed a collaborative model for home-based delivery of CGA-guided interventions. Methods: We performed a retrospective review of community-dwelling patients aged ≥ 70 years with an active cancer diagnosis who were deemed to be frail by a CGA and who received home-based CGA-guided supportive interventions under a collaborative care model established by the geriatric oncology and the palliative home care teams at Hartford HealthCare Cancer Institute between October 2020 and April 2022. The collaboration is based on multidisciplinary discussion of the CGA results / interventions and weekly rounds to discuss patient’s progress and emerging needs. Results: A total of 182 patients received a CGA during the study period, 54% (n = 99) were determined to be frail based on impairment in ≥ 7 geriatric domains. Among all eligible patients, 19 patients (19%) were included in the collaborative model. The median age of enrolled patients was 84 years (74-90), 12/19 (63%) had metastatic cancer, 7/19 (37%) received systemic treatment, and 8/19 (42%) received radiation. The average number of geriatric impairments per patients was 9.5 (7-12). Geriatric impairments addressed at home were skilled nursing (19/19), physical therapy (18/19), occupational therapy (12/19), speech-language-pathology (2/19), nutrition (5/19), or social worker (11/19) support at home. The average number of unplanned hospitalizations was 1.16 per patient (range 0-4), and the average unplanned emergency room visit was 0.89 per person (range 0-4) during study time. Transition to hospice occurred in 8/19 patients (42%), the median time to transition to hospice was 33.5 days (15-167). Conclusions: This study demonstrates the feasibility of a collaborative model for home-based geriatric oncology care in a real-world setting to help reduce the burden of care on patients and ensure patient-centered delivery of CGA-guided interventions. Findings underscore the need for future work to evaluate the impact of this novel geriatric oncology care model on patient outcomes.

Time toxicity in early phase clinical trials (EP-CTs)

236

Background: EP-CTs are an increasingly important treatment option for patients with cancer, yet often require intensive monitoring. Little is known about the time that EP-CT participants spend in-hospital, and how that time compares to study requirements. Methods: We retrospectively reviewed the electronic health record (EHR) of consecutive patients enrolled in EP-CTs at Massachusetts General Hospital from 2017-2019 to obtain patient characteristics (demographics and clinical factors) and EP-CT investigational agent (immunomodulatory therapy [IM], targeted inhibitor [TI], antibody drug conjugate [ADC]/chemotherapy prodrug). We identified protocol requirements by reviewing the study calendar for in-hospital days for any reason, including clinician visits and diagnostic tests. We identified the real-world number of days spent at the hospital by reviewing the EHR for in-hospital days. We used descriptive statistics to compare patient characteristics and outcomes for those with higher time toxicity, defined as 5+ real-world visits during the first 28 days on trial, versus lower time toxicity. Results: Among 421 patients (median age = 63.0 years, 56.9% female, 97.6% metastatic disease), 43.2% participated in IM trials, 43.0% TI, and 13.8% ADC. Most common tumor types were gastrointestinal (GI) (22.3%) and lung (20.0%). Over the first 28 days on trial, protocol requirements listed an average of 5.2 in-hospital days, yet real-world data demonstrated that patients had an average of 6.6 in-hospital days (p < 0.001). TI trial participants had the highest average number of anticipated protocol visits compared with those on other trials (5.5 [TI] vs 5.3 [ADC] vs 5.0 [IM], p = 0.027). In real-world data, those on ADC trials had the highest average number of visits (7.5 [ADC] vs 7.1 [TI] vs 5.7 [IM], p < 0.001). Those with 5+ real-world visits during the first 28 days were more likely to have GI cancer (25.8% vs 13.9%, p = 0.011) and less likely to have lung cancer (16.7% vs 27.9%, p = 0.011). Patients with more visits were also less likely to have traveled 50+ miles to the hospital (48.8% vs 59.8%, p = 0.04). Notably, 19.5% of patients (N = 82) were hospitalized during the first 28 days on trial, with an average length of stay of 4.9 days. Those with 5+ visits had fewer days, on average, from trial start to admission (371.9 vs 650.8, p < 0.001) and fewer days on trial (mean 156.0 vs 235.0, p = 0.001). There was no significant difference in days from trial start to death for those with higher versus lower time toxicity (mean 464.6 vs 526.4, p = 0.177). Conclusions: EP-CTs represent a potentially time-intensive treatment option, as we found that patients spend over one-fifth of their first 28 days on trial at the hospital for various visits. Our findings indicate that patients may often experience more in-hospital days than what the protocol states. These data could help inform patient-clinician discussions regarding EP-CT participation and the potential time toxicity involved.

Trends in the improvement in survival among patients with diffuse large B-cell lymphoma (DLBCL)

147

Background: Advances in treatment for patients with DLBCL have led to improved patient outcomes, but whether disparities exist with regards to improved survival outcomes remains understudied. We sought to describe changes in DLBCL survival rates over time and explore potential differential survival patterns by patients’ race and age. Methods: We utilized the Surveillance, Epidemiology, and End Results (SEER) database to identify patients diagnosed with DLBCL from 1980-2009 and determined 5-year survival rates for all patients, categorizing patients by year of diagnosis. We used descriptive statistics and logistic regression, adjusting for stage and year of diagnosis, to describe changes in 5-year survival rates over time by race/ethnicity and age. Results: We identified 43,564 patients with DLBCL eligible for this study (median age 67 years [ages: 18-64 = 44.2%, 65-79 = 37.1%, 80+ = 18.7%], 53% male, 40% stage III/IV). Most were White race (81.4%), followed by race/ethnic groups of Asian/Pacific Islander (API) (6.3%), Black (6.3%), Hispanic (5.4%), and American Indian/Alaska Native (AIAN) (0.05%). Overall, the 5-year survival rate improved from 35.1% in 1980 to 52.4% in 2009 across all patients (odds ratio (OR) for 5-year survival with increasing year of diagnosis = 1.05, p <.001). Patients in race/ethnic minority groups (API: OR = 0.86, p <.0001; Black: OR = 0.57, p <.0001; AIAN: OR = 0.51, p =.008) and older adults (ages 65-79: OR = 0.43, p <.0001; ages 80+: OR = 0.13, p <.0001) had worse 5-year survival rates adjusting for race, age, stage, and diagnosis year. We found consistent improvement in the odds of 5-year survival for year of diagnosis across all race and ethnicity groups (White: OR = 1.05, p <.001; API: OR = 1.04, p <.001; Black: OR = 1.06, p <.001; Hispanic: OR = 1.05, p <.005; AIAN: OR = 1.05, p <.001) and age groups (ages 18-64: OR = 1.06, p <.001; ages 65-79: OR = 1.04, p <.001; ages 80+: OR = 1.04, p <.001). Conclusions: In this analysis of SEER data, we found that patients with DLBCL experienced improvements in 5-year survival rates from 1980 to 2009, despite ongoing disparities demonstrating worse survival among patients in race/ethnic minority groups and older adults. Notably, we did not find differential improvement in the odds of 5-year survival within subgroups of race and age, underscoring the need for ongoing investigations to address disparities in cancer care delivery and outcomes.

Patient-reported hope, quality of life, symptom burden, coping, and financial toxicity in early-phase clinical trial participants

275

Background: Early phase clinical trials (EP-CTs) investigate novel treatment options in oncology, with recent advances in personalized therapy leading to improved outcomes and offering hope to patients with cancer. However, little research has sought to understand associations of patient-reported hope with quality of life (QOL), symptom burden, coping, and financial toxicity in EP-CT participants. Methods: We prospectively enrolled consecutive adults with cancer participating in EP-CTs at Massachusetts General Hospital from 04/2021-05/2022. Participants completed baseline surveys prior to treatment initiation that assessed hope (Herth Hope Index [HHI], higher scores indicate greater hope), QOL (Functional Assessment of Cancer Therapy-General), symptom burden (physical: Edmonton Symptom Assessment System [ESAS]; psychological: Patient Health Questionaire-4 [PHQ4]), coping (Brief COPE: self-blame, acceptance, denial, emotional support, active, behavioral disengagement), and financial toxicity (COST tool, higher scores indicate greater financial wellbeing). We used regression models to determine associations of hope scores with patient-reported QOL, symptom burden, coping, and financial toxicity. Results: Of 157 eligible patients, we enrolled 129 (enrollment rate 82.2%, median age = 62.5 years [range 33.0-83.0], 53.9% female, and 96.0% metastatic cancer). Most common cancer types were gastrointestinal (37.5%), breast (20.3%), lung (8.6%), and head and neck (7.8%). Patients had an average HHI score of 27.5 (range 15.3 – 36.0), with 30.5% reporting high levels of hope. We found associations of higher hope scores with better QOL (B = 0.24, p < 0.001) and lower symptom burden (ESAS-physical: B = -0.14, p < 0.001; PHQ4-depression: B = -2.07, p < 0.001; PHQ4-anxiety: B = -0.93, p = 0.001). We also found that hope scores were associated with patients’ coping (self-blame [B = -1.44, p < 0.001]; acceptance [B = 1.40, p < 0.001], denial [B = -1.12, p = 0.004], emotional support [B = 0.99, p < 0.001], active [B = 1.02. p = 0.001], behavioral disengagement [B = -2.52, p < 0.001]). Lastly, we found that higher hope scores were associated with greater financial wellbeing (B = 0.11, p = 0.026). Conclusions: In this prospective cohort study, we demonstrated a substantial proportion of EP-CT participants had high baseline hope and identified associations of hope scores with other important patient-reported outcomes. Specifically, we found novel associations of higher hope scores with better QOL, lower symptom burden, more adaptive coping mechanisms, and greater financial wellbeing, underscoring the importance of targeting these patient-reported outcomes when seeking to enhance the care experience of EP-CT participants.

Supportive oncology care at home after discharge for patients with advanced cancer

295

Background: Patients with advanced cancer often experience frequent and prolonged hospitalizations, and the transition from hospital to home represents a critical period for these individuals, as they prefer to maximize time at home and avoid readmissions. We sought to demonstrate the feasibility and acceptability of a Supportive Oncology Care at Home intervention to address the post-discharge needs of recently hospitalized patients with advanced cancer. Methods: We conducted a single-arm pilot trial at Massachusetts General Hospital (MGH). We enrolled English-speaking adults with advanced solid tumors experiencing their second or later unplanned hospitalization, who were being discharged home without hospice services and residing within a 50 mile radius of MGH. The three-week intervention consisted of: 1) hospital in the home care model for proactive symptom assessment and management, including clinician visits to assess patients, draw labs, administer intravenous medications and hydration, and ensure optimal symptom management; 2) remote monitoring of daily patient-reported symptoms, vital signs, and body weight; and 3) structured communication with the oncology team. The primary endpoint of the study was feasibility, defined as ≥60% of approached and eligible patients enrolling and ≥60% of participants completing daily symptom assessments. After intervention completion, patients rated the helpfulness and convenience of the intervention and symptom monitoring technology. Results: From 12/2021-6/2022, we enrolled 40 out of 66 approached patients (60.6% enrollment rate). Enrolled patients (median age = 58.5 years, 50% female, 75% white, 68% married, 50% gastrointestinal cancers) completed 93.8% of daily symptom assessments. 12 patients (30%) did not complete the intervention due to withdrawal (5), hospice transfer (4), or death (3). Among enrolled patients, 20.0% were enrolled in hospice and 15.4% died at 30 days after hospital discharge. In exit interviews, 100% and 75% rated the intervention and symptom monitoring as helpful, respectively. 83% of patients found the in-home monitoring technology convenient. Conclusions: We found that a three-week Supportive Oncology Care at Home intervention is a feasible approach to providing post-discharge care for high acuity, seriously ill hospitalized patients with advanced cancer. These patients also found the intervention highly acceptable. Future studies will test the efficacy of the intervention for reducing hospital readmissions, improving symptom management and quality of life, and increasing days spent at home near the end of life. Clinical trial information: NCT04637035.

Clinical trial perceptions among patients with gastrointestinal (GI) cancer in an academic cancer center

e18579

Background: Clinical trials (CTs) are essential for advancing care for individuals with cancer, yet a minority of patients participate in CTs. We conducted an exploratory analysis of CT perceptions (knowledge, attitudes, and barriers) among patients with GI cancer. Methods: We prospectively surveyed a convenience sample of patients treated for GI cancer at Massachusetts General Hospital from 11/2020–12/2021. We obtained sociodemographic/clinical characteristics via patient report and chart review. We assessed CT perceptions, communication confidence (Perceived Efficacy in Patient-Physician Interactions Questionnaire, PEPPI, range 0-50; higher score indicates higher confidence), and illness perceptions (Brief Illness Perceptions Questionnaire, BIPQ, range 0-80; higher score indicates more negative perceptions). Using descriptive statistics, we examined associations of CT perceptions with patient characteristics, communication confidence, and illness perceptions. Results: We enrolled 80 patients (median age = 66 years [range 24-85], 54% men, 93% white, 48% had metastatic disease, 69% had college/postgraduate degrees, 15% had participated in CTs [25% phase I, 75% phase II-III]). Cancer types were pancreatic (38%), colorectal/bowel (23%), hepatobiliary (18%), gastroesophageal (16%), and other (6%). Median PEPPI was 43 (range 16-50). Median BIPQ was 44 (range 1-66); 63% reported negative illness perceptions (BIPQ≥40). Most (89%) agreed CTs are essential to improving standard treatment, yet only 42% had discussed CTs during care. Overall, 38% reported a clear idea of what a CT means, and 8% thought most people are cured on CTs. Unmarried patients were more likely to think most people are cured on CTs (21% vs 4%, p = .017). Most (76%) saw CTs as opportunities to obtain new treatment; only 15% believed enrolling would mean missing out on standard care. Only 16% thought most patients like them enroll in CTs. Younger patients (≤65 years) were more likely to think most patients like them enroll (25% vs 8%, p = .046). Most (61%) felt confident differentiating a CT from other treatments. Older patients (> 65 years) were more likely than younger patients to agree/strongly agree that they feel confident differentiating a CT from other treatments (74% vs 49%, p = .025), as were men versus women (76% vs 43%, p = .003). Those with higher communication confidence (Medians [M]: 44 vs 40, p = .04) or more positive illness perceptions (M: 41 vs 50, p = .003) were also more likely to be confident differentiating a CT. Conclusions: In this exploratory analysis of CT perceptions among patients with GI cancer, we found high levels of CT knowledge and positive CT perceptions. We demonstrated hypothesis-generating associations among patient factors and CT perceptions, underscoring the need for future research to confirm our findings and to develop interventions to enhance CT decision-making and participation.

Relationship of travel distance with patient demographics, advance care planning, and survival in early-phase clinical trials (EP-CTs)

6558

Background: EP-CTs are often conducted at large academic centers, which may require some patients to travel further for their care. Little is known about either the distance EP-CT participants travel for their care or the association of distance traveled with patient characteristics and outcomes. Methods: We retrospectively reviewed the electronic health records of consecutive patients enrolled in EP-CTs at Massachusetts General Hospital from 2017-2019 to obtain patient characteristics (demographics and clinical factors) and outcomes (including time spent on trial, survival, and presence or absence of an advance care planning [ACP] discussion, defined as documentation of a code status or goals of care conversation in the medical record). We also used patients’ home zip codes to derive the social deprivation index (SDI; a composite demographic measurement from 0-100 quantifying social determinants of health, with higher numbers indicating more disadvantage). To estimate distance traveled, we calculated the miles traveled in one direction driving from home zip code to trial site. We used descriptive statistics to compare patient characteristics and outcomes for those traveling < 50 miles (short distance) versus those traveling 50+ miles (long distance). Results: Among 421 patients (median age = 63.0 years, 56.9% female, 97.6% metastatic disease), median distance traveled was 36.4 miles. Half of patients (n = 217; 51.5%) traveled 50+ miles to receive care on trial. There were no significant differences between those traveling short and long distances in most patient characteristics evaluated, including age (60.9 vs 60.6 years; p = 0.635), sex (53.9% female vs 57.6%; p = 0.447), race (85.3% white vs 84.8%; p = 0.346), marital status (71.8% married vs 69.3%; p = 0.586), insurance (51% private vs 54.4%; p = 0.266), cancer type (22.5% GI vs 21.2%; p = 0.666), prior lines of therapy (52.5% one-two lines vs 51.2%; p = 0.981), and performance status (62.3% ECOG 1 vs 66.8%; p = 0.270. However, those with a higher SDI score were less likely to travel a long distance for trial participation (mean SDI 36.7 for short distance vs 30.5 for long distance; p = 0.026). Patients traveling a long distance were less likely to have a documented ACP discussion (48.8% vs 66.7%; p < 0.001). We found no significant difference in time spent on trial between those traveling short and long distances (mean days: 98 vs 93.5; p = 0.175) or in time from coming off trial to death (mean days: 147.7 vs 153.7; p = 0.099). Conclusions: We found that half of EP-CT participants travel 50+ miles in one direction to their trial site, with disparities in travel distance based on the social deprivation index. Notably, those traveling long distances were less likely to have a documented ACP discussion. Our findings suggest several unmet needs in the EP-CT population and highlight opportunities for future intervention development.

Protocol requirements and logistical intensity of early-phase clinical trials (EP-CTs)

e18609

Background: EP-CTs are increasingly important options for patients with cancer and often involve intensive monitoring. Characterizing the time burden and logistical intensity of EP-CT protocols could help patients and clinicians make informed decisions about trial participation. Methods: We retrospectively reviewed the electronic health records of consecutive patients enrolled in EP-CTs at Massachusetts General Hospital from 2017-2019 to obtain baseline characteristics (demographics and clinical factors), EP-CT investigational agent (immunomodulatory [IM], targeted inhibitor [TI], antibody drug conjugate [ADC]/chemotherapy prodrug), and logistical intensity (visit frequency required per protocol and presence of extended visits). We defined visit frequency as the number of visits required per protocol within the first 28 days on trial. We defined an extended visit as six or more hours required in clinic on at least one day during the first 28 days on study. We evaluated associations among patient characteristics, investigational agents, logistical intensity, and time spent on trial. Results: Among 421 patients (median age = 63.0 years, 56.9% female, 97.6% metastatic disease), 43.2% were enrolled in IM EP-CTs, 43.0% TI, and 13.8% ADC/chemotherapy prodrug investigational agents. Patients enrolled on ADC/prodrug trials had the highest burden of metastatic disease (mean sites: 2.8 [ADC] vs 2.4 [TI] vs 2.3 [IM], p = 0.007) and oldest age (mean years: 64.0 [ADC] vs 61.7 [IM] vs 58.5 [TI], p = 0.003). However, those on ADC trials had the most days spent on trial (mean days: 78.3 [TI] vs 102.2 [IM] vs 131.8 [ADC], p = 0.003). Patients enrolled on TI trials had the highest required visit frequency compared with those enrolled on other trials (mean visits: 5.5 [TI] vs 5.3 [ADC] vs 5.0 [IM], p = 0.027). Additionally, those on TI trials were most likely to have an extended visit (82.3% [TI] vs 58.2% [IM] vs 29.3% [ADC], p < 0.001) and least likely to receive first in human therapy (38.1% [TI] vs 74.1% [ADC] vs 74.2% [IM], p < 0.001). Conclusions: In this cohort of patients participating in EP-CTs, we found that those enrolled on TI trials had the highest per protocol visit frequency and greatest likelihood of required extended visits. Those on ADC trials spent the most days on trial despite having the highest average age and burden of metastatic disease. These data highlight the time burden and logistical intensity of EP-CTs, underscoring certain trials as especially time intensive, which may help inform trial selection and participation.

Patient-reported hope, quality of life (QOL), symptom burden, and coping mechanisms in early phase clinical trial participants

12114

Background: Early phase clinical trials (EP-CTs) investigate novel treatment options in oncology, with recent advances in personalized therapy leading to improved outcomes and offering hope to patients with cancer. However, little research has sought to understand associations of patient-reported hope with QOL, symptom burden, and coping mechanisms in EP-CT participants. Methods: We prospectively enrolled consecutive adults with cancer participating in EP-CTs at Massachusetts General Hospital from 04/2021-01/2022. Participants completed baseline surveys prior to treatment initiation that assessed hope (Herth Hope Index [HHI], higher scores indicate greater hope), QOL (Functional Assessment of Cancer Therapy-General), symptom burden (physical: Edmonton Symptom Assessment System [ESAS]; psychological: Patient Health Questionaire-4 [PHQ4]), and coping mechanisms (Brief COPE). We used independent samples t-test to test for mean differences between groups and regression models to explore associations of hope with patient characteristics as well as patient-reported QOL, symptom burden, and coping mechanisms. Results: Of 92 eligible patients, we enrolled 85 (enrollment rate 92.4%, median age = 61.4 years [range 54.7-68.9]; 56.5% female, and 95.3% metastatic cancer). Most common cancer types were gastrointestinal (41.2%), breast (21.2%), lung (7.1%), and gynecologic (7.1%). Patients had an average HHI score of 28.2 (range 12.0-36.0), with 32.9% reporting high levels of hope. We found that married patients had higher mean hope score compared with non-married patients (28.9 versus 26.1, p = 0.024), those with children had higher mean hope scores than those without (28.9 versus 25.9, p = 0.013), and those who had received 3 or more lines of prior therapy compared with 1-2 (29.3 versus 27.2, p = 0.045) had higher hope scores. We also found associations of hope with patients’ QOL (B = 0.24, p < 0.001), symptom burden (ESAS-physical: B = -0.13, p = 0.001; PHQ4-depression: B = -2.26 p = < 0.001; PHQ4-anxiety: B = -0.94, p = 0.008), and coping (self-blame [B = -1.39, p = 0.003]; acceptance [B = 1.23, p = 0.002], denial [B = -1.09, p = 0.009], support [B = 1.06, p = 0.002], active [B = 0.73. p = 0.034], disengage [B = -3.24, p < 0.001]). Conclusions: In this prospective cohort study, we demonstrated that a substantial proportion of EP-CT participants had high baseline hope, and we identified several patient factors associated with their hope scores. We also found novel associations of higher hope scores with better QOL, lower symptom burden, and more adaptive coping mechanisms. Collectively, our findings highlight the potential for patient-reported hope to represent a key factor to consider when seeking to improve outcomes in EP-CT participants.

CONTINUUM: A pilot care transition intervention for hospitalized patients with advanced cancer

1534

Background: Patients with advanced cancer are frequently hospitalized and experience burdensome transitions of care after discharge. Interventions to address patients’ symptoms, support medication management, and ensure continuity of care after discharge are lacking. We sought to demonstrate the feasibility and acceptability of CONTINUUM (CONTINUity of care Under Management by video visits) for this population. Methods: We conducted a single-arm pilot trial (n = 54) of CONTINUUM at Massachusetts General Hospital (MGH). The intervention consisted of a video visit with an oncology nurse practitioner (NP) within 3 business days of hospital discharge to address symptoms, medication management, hospitalization-related issues, and care coordination. Prior to discharge, we enrolled English-speaking adults with advanced breast, gastrointestinal, genitourinary, or thoracic cancers experiencing an unplanned hospitalization who were receiving ongoing oncology care at MGH and being discharged home without hospice services. We defined the intervention as feasible if ≥70% of approached and eligible patients enrolled and if ≥70% of enrolled patients completed the intervention within 3 business days of discharge. At 2 weeks after discharge, patients rated the ease of use of the video technology and stated whether they would recommend the intervention. NPs completed post-intervention surveys to assess fidelity to the intervention protocol. Results: From 01/07/21 to 05/28/21, we enrolled 54 patients (77.3% of patients approached). Of the enrolled patients (median age = 65.0 years; 59.3% and 22.2% had advanced gastrointestinal or thoracic cancers, respectively), 83.3% of enrolled patients received the intervention within 3 business days of discharge. Patient rating of the ease of use of video technology was a mean of 7.8 out of 10, with 71.4% stating they “agreed” or “strongly agreed” that they would recommend the intervention. NP post-intervention surveys revealed that visits focused on symptom management (85.7%), followed by addressing post-hospital care issues (69.0%). At 30 day follow up, 38.8% were readmitted within 30 days of discharge, and 12.2% died within 30 days of discharge. Conclusions: We found that CONTINUUM, which consists of an NP-delivered video visit soon after hospital discharge addressing patients’ symptoms, medications, and care coordination, represents a feasible and acceptable approach to provide post-discharge care for hospitalized patients with advanced cancer. Future studies will test the efficacy of the intervention for reducing hospital readmissions. Clinical trial information: NCT04640714.

Associations of sarcopenia with hematologic toxicity, treatment intensity, and healthcare utilization in patients with metastatic colorectal cancer

84

Background: We evaluated the impact of baseline sarcopenia on hematologic toxicity, treatment intensity, and healthcare utilization in patients with mCRC receiving FOLFOX or FOLFIRI. Methods: We retrospectively analyzed patients with mCRC who received care at our institution from 1/2011-11/2018 and were part of a biobanking protocol. Included adults received either first-line palliative FOLFOX- or FOLFIRI-based regimens and were followed for 6 months. We categorized sarcopenia based on skeletal muscle index measured at diagnosis of metastatic disease and pre-defined sex-specific cutoff values (F < 39 cm2/m2, M < 55cm2/m2). Our primary aim was to evaluate the association of sarcopenia and hematologic toxicity, defined as the incidence of grade ≥3 (G≥3) neutropenia, thrombocytopenia, or anemia (NCI CTCAE v5.0). Secondary endpoints included treatment intensity (dose reductions, treatment delays, relative-dose intensity [RDI]), and healthcare utilization (ED visits and/or hospitalizations). Bivariate analyses were used to evaluate associations between baseline sarcopenia and outcomes. Results: 126 of 177 screened patients met inclusion criteria (70 (56%) males, median age 61 yrs (range, 29-85)). 59 (46.8%) patients were sarcopenic. More patients received FOLFOX than FOLFIRI (92 [73.0%] vs. 34 [27.0%]). At baseline, patients had a median weight 76.9kg (IQR, 70.0-90.4 kg), BMI 26.6 kg/m2 (IQR, 24.1-30.5 kg/m2), and BSA 1.90 m2 (IQR, 1.72-2.01 m2). The incidence of G≥3 hematologic toxicity was 39.0% vs. 23.9% in sarcopenic and non-sarcopenic patients, respectively (p = 0.06). Patients with sarcopenia experienced higher incidence of G≥3 neutropenia (30.5% vs. 14.9%, p = 0.03), while G≥3 thrombocytopenia was similar (3.4% vs. 1.5%). The incidence of dose reductions and treatment delays did not differ significantly (86.4% vs. 89.5%, 72.9% vs. 71.6%, respectively). RDI was decreased for the 5FU bolus (52.5% vs. 65.0%, p = 0.02). Rates of ED visits (32.2% vs. 19.4%, p = 0.10) and hospitalizations (32.2% vs. 26.9%, p = 0.51) did not differ compared between patients with and without sarcopenia. Conclusions: Patients with mCRC and baseline sarcopenia receiving FOLFOX- or FOLFIRI experienced a higher incidence of G≥3 neutropenia and lower 5FU bolus treatment intensity. Studies are needed to understand how best to adjust treatment according to patients’ muscle mass.

Patient-reported outcomes (PROs) in pancreatic cancer clinical trials (CTs)

534

Background: Incorporating PROs into CTs is critical for patients with pancreatic cancer, as these individuals often experience a high symptom burden and prioritize maximizing quality of life (QOL). Methods: We reviewed protocols of completed U.S. CTs listed on ClinicalTrials.gov that investigated curative/palliative interventions in pancreatic cancer from 1995-2020. We assessed up to August 2021 if CT publications reported PRO results through linked publications and an independent search through PubMed/MEDLINE. We extracted CT details (funding, eligibility, etc), and if PROs (outcome directly reported by patients) were listed as an endpoint (primary/secondary/exploratory). We classified interventions as cancer-directed (e.g. chemotherapy), supportive care (e.g. neurolysis), or other (e.g. curcumin). Results: We reviewed 619 protocols and included 379 in the analysis. Most CTs investigated cancer-directed interventions (317, 83.6%). Only 43 (11.4%) included PROs as endpoints (Table). In these, most of the PROs assessed QOL (34, 79.1%) and pain (15, 34.9%). For the 33/43 (76.7%) protocols that listed a specific PRO instrument, EORTC-QLQ C30 (11/33, 33.3%) was the most common. Only 6 (18.2%) protocols included pancreatic cancer-specific PROs, such as QLQ-PAN26. Supportive care CTs were more likely to assess PROs than cancer-directed CTs (odds ratio, OR= 62.6, 95% CI 16.7-234.3, p<0.0001). Protocols listed PROs as a primary, secondary, and exploratory endpoint in 15 (34.9%), 25 (58.1%), and 3 (6.9%) CTs respectively. Most CTs (13/15, 86.7%) with PROs as a primary endpoint evaluated supportive care interventions. Of 15 CTs with PROs as a primary endpoint, 10 (66.7%) had results published. Of 28 CTs assessing PROs as a secondary/exploratory outcome, 20 (71.4%) had published results and 12 (42.9%) included PRO data. Conclusions: From 1995-2020, only 11.4% of pancreatic cancer CTs incorporated PROs as endpoints. Supportive care CTs were more likely to include PROs than cancer-directed CTs. Our findings underscore the need to improve efforts to incorporate PROs into CTs for patients with pancreatic cancer. Data as number (%).[Table: see text]

CONTINUUM: A single-arm pilot care transition intervention for hospitalized patients with advanced cancer

33

Background: Patients with advanced cancer are frequently hospitalized and experience burdensome transitions of care after discharge. Interventions to address patients’ symptoms, support medication management, and ensure continuity of care after discharge are lacking. We sought to demonstrate the feasibility and acceptability of CONTINUUM (CONTINUity of care Under Management by video visits) for this population. Methods: We conducted a single-arm pilot trial (n = 50) of CONTINUUM at Massachusetts General Hospital (MGH). The intervention consisted of a video visit with an oncology nurse practitioner (NP) within 3 business days of hospital discharge to address symptoms, medication management, hospitalization-related issues, and care coordination. Prior to discharge, we enrolled English-speaking adults with advanced breast, gastrointestinal, genitourinary, or thoracic cancers experiencing an unplanned hospitalization who were receiving ongoing oncology care at MGH and being discharged home without hospice services. We defined the intervention as feasible if ≥70% of approached and eligible patients enrolled and if ≥70% of enrolled patients completed the intervention within 3 business days of discharge. At 2 weeks after discharge, patients rated the ease of use of the video technology and stated whether they would recommend the intervention. NPs completed post-intervention surveys to assess fidelity to the intervention protocol. Results: From 01/07/21 to 05/28/21, we enrolled 50 patients (75% of patients approached). Of the enrolled patients (median age = 65 years; 62% and 22% had advanced gastrointestinal or thoracic cancers, respectively), 78% of enrolled patients received the intervention within 3 business days of discharge. Patient rating of the ease of use of video technology was a mean of 7.6 out of 10, with 72% stating they “agreed” or “strongly agreed” that they would recommend the intervention. NP post-intervention surveys revealed that visits primarily focused on symptom management (56%), followed by addressing post-hospital care issues (21%). Of the 30 patients with 30-day follow-up, 43% were readmitted within 30 days of discharge, and 17% died within 30 days of discharge. Conclusions: We found that CONTINUUM, which consists of an NP-delivered video visit soon after hospital discharge addressing patients’ symptoms, medications, and care coordination, represents a feasible and acceptable approach to provide post-discharge care for hospitalized patients with advanced cancer. Future studies will test the efficacy of the intervention for reducing hospital readmissions. Clinical trial information: NCT04640714.

Types of information that patients with lung cancer with targetable driver mutations and their caregivers learn from online forums: Results of a qualitative study

161

Background: Patients and caregivers seek information about cancer from a variety of sources in addition to their medical team. In recent years, patients with specific lung cancer subtypes have created advocacy groups with active online forums for networking, support, and information exchange. We sought to describe the types of information that patients and caregivers learned from their peers online about lung cancer with targetable driver mutations. Methods: In this qualitative study, we enrolled patients with lung cancer with targetable driver alterations in EGFR, ALK, or ROS1 genes and their caregivers. We conducted individual, semi-structured interviews with participants from a single academic center, asking about their experiences learning about and living with lung cancer. Three reviewers used a framework analysis to code transcripts, and we synthesized the codes into themes. In this analysis, we focused on one theme that emerged from the data related to the information that participants learned from online forums. Results: Of 59 patients approached, 39 patients (mean age = 59, 56% female, median time from diagnosis 16 months) and 16 caregivers (69% male, all spouses or long-term partners) agreed to participate and completed interviews. Participants used online forums to compare their experiences with others to gain a preview of what might lie ahead. Specifically, they read about patients with long survival that were a source of hope, as well as patients’ experiences of progression and dying from cancer that made clinical estimates of prognosis more personal and vivid. Online forums provided a venue to learn about the latest research, available clinical trials, how to manage side effects, and where to find expert clinicians. Participants learned about the treatments received by patients at other centers. They were also exposed to emotionally intense stories of patients from around the world who lacked access to targeted therapy and sought advice from other forum members about how to find specialized care. Caregivers pointed out the distressing effect on patients of learning bad news about online peers as a downside of online forums, but overall found them beneficial and supported patients’ engagement with them. Conclusions: Online forums, particularly those that narrowly focus on specific diseases or treatments, provide patients and caregivers with anecdotal evidence that helps them with practical matters, such as how to manage side effects, as well as more existential issues, such as how long they can expect to live. These results suggest that clinicians should be open to and curious about the information that patients and caregivers learn from online forums in order to better understand the perspectives that patients and caregivers bring to discussions and decisions about their cancer.

Illness perceptions, financial toxicity, symptom burden, and survival in cancer clinical trial (CCT) participants

160

Background: Patients’ perceptions of their illness are important for treatment decision-making and quality of life. Limited data exist describing associations of illness perceptions with other patient-centered outcomes, particularly in CCT participants. We sought to examine associations among illness perceptions and CCT patients’ financial toxicity, physical and psychological symptoms, and survival. Methods: From 7/2015-7/2017, we prospectively enrolled CCT participants who expressed interest in financial assistance programs (n = 157) and a group of patients matched by age, sex, cancer type, specific trial, and trial phase (n = 103). We assessed baseline illness perceptions (Brief Illness Perceptions Questionnaire [BIPQ] with scores > 50 indicating negative perceptions), financial toxicity (degree costs of cancer care have been a burden, moderate to catastrophic indicating financial toxicity), physical (Edmonton Symptom Assessment Scale [ESAS]) and psychological (Patient Health Questionnaire-4 [PHQ-4]) symptoms. We used descriptive statistics to examine associations of BIPQ and sociodemographic/clinical factors, financial toxicity, ESAS, PHQ-4, and overall survival. We used the Kaplan-Meier method to estimate median survival times and Cox regression to assess the association of BIPQ and overall survival. Results: Among 260 patients, 189 (72.7%) completed BIPQ surveys (median age 69 [Range 26 to 83] years, 66.1% female). 68.8% had negative illness perceptions. We found no significant associations among negative illness perceptions and patients’ age, sex, race, education, marital status, performance status, insurance, cancer type, metastatic disease status, self-reported income, trial phase, trial year, or Charlson Comorbidity Index score. Patients with negative illness perceptions were more likely to report financial toxicity (69.8% vs 48.8%, p = 0.006), and had higher ESAS-total (Medians: 44 [Range 0-89] vs 21 [Range 0-78], p < 0.001), PHQ-4 depression (Medians: 2 [Range 0-6] vs 0 [Range 0-6], p < 0.001), and PHQ-4 anxiety (Medians: 3 [Range 0-6] vs 1 [Range 0-6], p < 0.001) scores. Patients with negative illness perceptions had shorter overall survival (Medians: 22 [Range 10-29] vs 42 [Range 28-Not Reached] months, log-rank p = 0.004). Adjusting for receipt of financial assistance, patients with negative illness perceptions experienced higher risk of death (HR 1.65, 95% CI 1.10-2.48). Conclusions: In this prospective study of CCT participants, we found that patients with negative illness perceptions experienced greater financial toxicity, more symptom burden, and worse survival than those with more positive perceptions, despite comparable sociodemographic/clinical factors. These findings highlight the need to assess and address patients’ illness perceptions and financial burden when seeking to enhance patient-centered outcomes in oncology.

Financial toxicity, symptom burden, illness perceptions, and communication confidence in cancer clinical trial participants

86

Background: Cancer clinical trial (CCT) participants are at risk for experiencing adverse effects from financial toxicity, yet these remain understudied. We sought to describe associations among CCT participant-reported financial toxicity (financial burden [FB] and trial cost concerns), symptoms, illness perceptions, communication confidence, hospitalizations, and survival. Methods: From 7/2015-7/2017, we prospectively enrolled CCT participants who expressed interest in financial assistance (n = 100) and a patient group matched by age, sex, cancer type, specific trial, and trial phase (n = 98). We assessed FB (burdened by costs of cancer care), trial cost concerns (worried about affording medical costs of a CCT), physical (Edmonton Symptom Assessment Scale [ESAS]) and psychological (Patient Health Questionnaire-4 [PHQ-4]) symptoms, illness perceptions (Brief Illness Perception Questionnaire [BIPQ]), and communication confidence (Perceived Efficacy in Patient-Physician Interactions [PEPPI]). We used regression models to explore sociodemographic associations with FB and trial cost concerns, as well as their associations with symptom burden, illness perceptions, and communication confidence, adjusting for age, sex, race, performance status, marital status, and metastatic status. We also used Kaplan-Meier and regression methods to evaluate their associations with 6-month hospitalizations and survival. Results: Of 198 patients enrolled, 112 (56.6%) reported FB and 82 (41.4%) had trial cost concerns. Patients with FB were younger (OR 0.96, 95% CI 0.94-0.98) with lower incomes (< $100,000, OR 4.61, 95% CI 2.35-9.01). Patients with trial cost concerns had lower incomes (< $100,000, OR 2.78, 95% CI 1.45-5.29). On adjusted analyses, patients with FB had higher ESAS (OR 1.03, 95% CI 1.02-1.05), PHQ-4 depression (OR 1.54, 95% CI 1.22-1.94), and PHQ-4 anxiety (OR 1.30, 95% CI 1.08-1.55) scores, as well as more negative illness perceptions (OR 1.04, 95% CI 1.01-1.07), but no significant difference in communication confidence (OR 0.98, 95% CI 0.93-1.05). Patients reporting trial cost concerns had higher ESAS (OR 1.03, 95% CI 1.01-1.05), PHQ-4 depression (OR 1.35, 95% CI 1.10-1.65), and PHQ-4 anxiety (OR 1.27, 95% CI 1.07-1.51) scores, as well as more negative illness perceptions (OR 1.06, 95% CI 1.03-1.10), and lower communication confidence (OR 0.93, 95% CI 0.87-0.99). Financial toxicity was not significantly associated with hospitalizations or survival. Conclusions: In this study of CCT participants, younger patients with lower incomes were most vulnerable to financial toxicity. Financial toxicity was associated with greater symptoms, more negative illness perceptions, and lower communication confidence, underscoring the importance of addressing these issues when seeking to alleviate the adverse effects of financial toxicity in CCT participants.

Limited benefit and high financial burden of drugs used to manage cancer-associated anorexia/cachexia syndrome (CACS)

55

Background: To date, drugs used for CACS have limited efficacy and may have unwanted side effects. Despite this, clinicians commonly prescribe certain drugs for CACS, adding financial burden to patients through out-of-pocket costs near or at the end of life. We sought to describe the range of costs of the multiple drugs commonly used to manage CACS. Methods: We reviewed oncology, nutrition, and supportive care guidelines (ASCO, ESMO, ESPEN, MASCC) and compiled a list of drugs recommended for CACS. We also included drugs commonly used off-label in clinical practice as identified by study authors. Using GoodRx.com, we extracted available formulations (e.g., tablet [tab] vs capsule [cap] vs oral disintegrating tablet [ODT] vs solution [sol]), and brand name vs generic products for each drug. We identified the average retail price, and the lowest price with coupons, for each formulation using the most commonly used dosage for a 2-week period (typical fill). The average retail price is the cash price for consumers without prescription drug coverage. The lowest price represents a discounted, ‘’best case’’ scenario of out-of-pocket costs for patients without prescription drug coverage. We collected data using the zip code 10065 in May 2021 to describe the range of costs of these drug formulations. Results: We included 7 drugs available in 20 formulations (Table). For a 2-week fill, the least expensive options included generic olanzapine 5 mg tab (once daily) and generic mirtazapine 15 mg tab (once daily), costing $4.50 per fill. Brand-name solutions were costly, ranging from $605.50 (megestrol acetate, Megace ES) to $1,156.30 (dronabinol, Syndros). Costs between formulations of the same drug/dosage varied widely: for olanzapine 5 mg, the lowest price varied from $4.50 [generic, tab] to $238.40 [Zyprexa zydis, brand-name, ODT]. Conclusions: We found that the costs of drugs used for CACS are highly variable, ranging from < $5 to > $1000 per fill. Overall there is limited data supporting use of drugs for managing CACS. Thus, our findings can help guide patient-clinician discussions about the risk/benefit ratio of a prescription for managing CACS, while highlighting the importance of seeking less expensive formulations when possible.[Table: see text]

Supportive oncology care at home intervention for patients with pancreatic cancer

155

Background: Patients with pancreatic cancer receiving chemotherapy often experience substantial symptoms and high healthcare utilization. We sought to determine the feasibility of delivering a Supportive Oncology Care at Home intervention designed to address the needs of patients receiving treatment for pancreatic cancer. Methods: We prospectively enrolled patients with pancreatic cancer who were participating in a parent trial of neoadjuvant FOLFIRINOX and residing in-state, within 50 miles of our hospital. Patients received the Supportive Oncology Care at Home intervention during neoadjuvant treatment (i.e., up to 4 months). The intervention entailed: 1) remote monitoring of daily patient-reported symptoms, daily vital signs, and weekly body weight; 2) a hospital in the home care model for symptom assessment and management; and 3) structured communication with the oncology team. We defined the intervention as feasible if ≥60% of patients enrolled in the study and ≥60% completed the daily assessments within the first two weeks of enrollment. We tracked numbers of phone calls, emails, and home visits generated by the intervention. We conducted exit interviews with patients, caregivers, and oncology clinicians to assess the acceptability of the intervention. We also compared rates of treatment delays, urgent clinic visits, emergency room (ER) visits, and hospitalizations among those who did (n = 20) and did not (n = 24) receive Supportive Oncology Care at Home from the parent trial. Results: From 1/2019-9/2020, we enrolled 80.8% (21/26) of potentially eligible patients. One patient became ineligible following consent due to moving out-of-state, resulting in 20 participants (median age = 67 years [range 55-77]; 60.0% female). In the first two weeks of enrollment, 65.0% of participants completed all daily assessments. Overall, patients reported 96.1% of daily symptoms, 96.1% of daily vital signs, and 92.5% of weekly body weights. Each participant generated an average of 2.22 phone calls (range 0.62-3.77), 2.96 emails (range 1.50-5.88), and 0.15 home visits (range 0-0.69) per week. During exit interviews, > 80% of patients, caregivers, and clinicians found the intervention to be helpful and convenient, and they reported high satisfaction with the communication among patients, clinicians, and the hospital in the home team. Patients receiving the intervention had lower rates of treatment delays (55.0% v 75.0%), urgent clinic visits (10.0% v 25.0%), ER visits or hospitalizations (45.0% v 62.5%), as well as a lower proportion of days spent in urgent clinic, ER, or hospital (2.7% v 7.8%), compared with those not receiving the intervention who were in the same parent trial. Conclusions: These findings demonstrate the feasibility and acceptability of a Supportive Oncology Care at Home intervention. Future work will investigate the efficacy of this intervention for decreasing healthcare use and improving patient outcomes. Clinical trial information: NCT03798769.

Changes in patient-reported outcomes (PROs) and tumor markers (TMs) to predict treatment response and survival in patients with metastatic gastrointestinal (GI) cancer

154

Background: PROs assessing quality of life (QOL) and symptoms at a single timepoint frequently correlate with clinical outcomes in patients with cancer, yet efforts to understand how longitudinal changes in PROs can predict for treatment outcomes are lacking. In practice, oncologists often use changes in serum TMs (CEA and CA19-9) to monitor patients with GI cancer, and thus we sought to examine associations of 1-month changes in PROs and TMs with treatment response and survival outcomes among patients with advanced GI cancer. Methods: We prospectively enrolled patients with metastatic GI cancer prior to initiating chemotherapy at Massachusetts General Hospital from 5/2019-12/2020. At baseline (start of treatment) and 1-month later, we collected PROs (QOL [Functional Assessment of Cancer Therapy General {FACT-G}], physical symptoms [Edmonton Symptom Assessment System {ESAS}], and psychological symptoms [Patient Health Questionnaire-4 {PHQ-4}]) and TMs. We used regression models to examine associations of 1-month changes in PROs and TMs with treatment response (clinical benefit [defined as decreased or stable tumor burden] or progressive disease at the time of first scan) and survival outcomes (progression-free survival [PFS] and overall survival [OS]), adjusted for baseline values of each respective variable. Results: We enrolled 159 of 191 patients approached (83.2% enrollment); 134 had 1-month follow-up data (median age = 64 years [range: 28 to 84 years], 64.2% male, 46.3% pancreaticobiliary cancer). For treatment response, 63.4% had clinical benefit and 36.6% had progressive disease at the time of first scan (mean time to first scan = 2.01 months). Changes in PROs (ESAS-Total: OR = 0.97, p = 0.022; ESAS-Physical: OR = 0.96, p = 0.027; PHQ-4 depression: OR = 0.67, p = 0.014; FACT-G: OR = 1.07, p = 0.001), but not TMs (CEA: OR = 1.00, p = 0.836 and CA19-9: OR = 1.00, p = 0.796), were associated with clinical benefit at the time of first scan. Changes in ESAS-Total (HR = 1.03, p = 0.004), ESAS-Physical (HR = 1.03, p = 0.021), PHQ-4 depression (HR = 1.22, p = 0.042), FACT-G (HR = 0.97, p = 0.003), and CEA (HR = 1.00, p = 0.001) were predictors of PFS. Changes in ESAS-Total (HR = 1.03, p = 0.006) and ESAS-Physical (HR = 1.04, p = 0.015) were predictors of OS, but 1-month changes in TMs (CEA: HR = 1.00, p = 0.377 and CA19-9: HR = 1.00, p = 0.367) did not significantly predict for OS. Conclusions: We found that 1-month changes in PROs can predict for treatment response and survival outcomes in patients with advanced GI cancers. Notably, 1-month changes in CEA only correlated with PFS, while changes in CA19-9 did not significantly predict treatment response or survival outcomes. These findings highlight the potential for early changes in PROs to predict treatment outcomes while underscoring the need to monitor and address PROs in patients with advanced cancer. Clinical trial information: NCT04776837.

Time burden and logistical intensity of early-phase clinical trials (EP-CTs)

84

Background: EP-CTs are increasingly important options for patients with cancer and often involve intensive monitoring. Thus, characterizing the time burden and logistical intensity of EP-CTs could help patients and clinicians make informed decisions regarding trial participation. Methods: We retrospectively reviewed the electronic health records of consecutive patients enrolled in EP-CTs at Massachusetts General Hospital from 2017-2019 to obtain baseline characteristics (demographics and clinical factors), EP-CT investigational agent (immunomodulatory therapy [IM], targeted inhibitor(s) [TI], antibody drug conjugate [ADC]/chemotherapy prodrug), and logistical intensity (trial visit frequency, presence of extended visits, distance traveled in one direction from home zip code to trial site). We defined visit frequency as the number of visits per protocol within the first 28 days on trial. We defined an extended visit as six or more hours in clinic on at least one day during the first 28 days on study. We investigated associations among patient characteristics, investigational agent, and logistical intensity. Results: Among 421 patients (median age=60.6 years, 55.8% female, 97.4% metastatic disease), most (73.6%) had two or more sites of metastatic disease. EP-CTs included 43.2% IM, 43.0% TI, and 13.8% ADC/chemotherapy prodrug. Patients enrolled in ADC/prodrug trials had the highest burden of metastatic disease (mean sites: 2.8 [ADC] vs 2.4 [TI] vs 2.3 [IM], p = 0.007) and oldest age (mean years: 64.0 [ADC] vs 61.7 [IM] vs 58.5 [TI], p = 0.003). Patients enrolled on TI trials had the highest visit frequency compared with those enrolled on other trials (mean visits: 5.5 [TI] vs 5.3 [ADC] vs 5.0 [IM], p = 0.027) and the fewest days spent on trial (mean days: 78.3 [TI] vs 102.2 [IM] vs 131.8 [ADC], p = 0.003). Patients enrolled on TI trials were also most likely to have an extended visit (82.3% [TI] vs 58.2% [IM] vs 29.3% [ADC], p < 0.001) and least likely to receive first in human therapy (38.1% [TI] vs 74.1% [ADC] vs 74.2% [IM], p < 0.001). Distance traveled from home to clinic did not significantly differ across trial type (median miles traveled: 35.1 [TI] vs 34.1 [IM] vs 33.2 [ADC], p = 0.884). Conclusions: In this cohort of patients participating in EP-CTs, we found that a plurality enrolled in IM studies. Those receiving ADC/prodrug regimens were older and had a higher burden of disease. On average, patients participating in EP-CTs had over five visits in the first month, with those enrolled on TI trials having the highest visit frequency and greatest likelihood of extended visits. Patients on TI trials also spent the fewest total days on trial. Despite the lack of significant differences in distance traveled, most patients were still traveling over 30 miles to get to the trial site. These data highlight the time burden and logistical intensity of various EP-CTs, which may help inform patient-clinician discussions about trial participation.

Identifying early-phase clinical trial (EP-CT) participants at risk for poor outcomes

301

Background: EP-CTs investigate novel treatment options, with recent advances in personalized therapy leading to increased response rates, decreased toxicity, and improved survival. Identifying EP-CT participants at risk for poor outcomes could help identify those who may benefit most from targeted supportive care efforts. Methods: We retrospectively reviewed the electronic health records of consecutive patients enrolled in EP-CTs from 2017-2019 to obtain baseline characteristics (demographics and clinical factors), clinical outcomes (survival, time on trial, completion of dose-limiting toxicity [DLT] period, emergency room [ER] visits, hospitalizations, and hospice use), and receipt of supportive care services before/during trial (palliative care, social work, physical therapy [PT], and nutrition). We calculated the validated Royal Marsden Hospital (RMH) prognosis score using data at the time of EP-CT enrollment based on patients’ lactate dehydrogenase, serum albumin, and number of sites of metastasis. RMH scores range from 0-3, with scores of 2+ indicating a poor prognosis. We examined differences in patient characteristics, clinical outcomes, and receipt of supportive care services based on the RMH prognosis score. Results: Among 350 patients (median age = 63.2 years [range 23.0-84.3]; 57.1% female, 98.0% metastatic cancer), the most common cancer types were lung (23.4%), gastrointestinal (20.3%), and breast (12.0%). Nearly one-third (31.7%) had an RMH score indicating a poor prognosis. Patients with a poor prognosis RMH score had a worse performance status (ECOG ≥1: 80.2% vs 58.1%, p <.001) and more prior treatment (3+ prior lines: 48.6% vs 34.7%, p =.001) than those with a better prognosis score. Those with a poor prognosis RMH score had worse survival (median: 147 vs 402 days, p <.001) and shorter time on trial (median: 49 vs 84 days, HR = 1.53, p <.001), as well as a lower likelihood of completing the DLT period (72.1% vs 80.8%, p =.015). Patients with a poor prognosis score had a higher risk for ER visits (HR 1.66; p =.037) and hospitalizations (HR 1.69; p =.016) while on trial, with earlier hospice enrollment (HR 2.22; p =.006) following the trial. Patients with a poor prognosis score were significantly more likely to receive palliative care before/during trial (46.8% vs 27.6% p =.001), but not social work (41.4% vs 41.4% p = 1.00), PT (44.1% vs 34.7%; p =.098), or nutrition (40.5% vs 37.2%; p =.557). Conclusions: EP-CT participants represent a unique population of patients with advanced cancer, and we identified a group at risk for particularly poor outcomes, including worse survival, shorter time on trial, and greater use of healthcare services. Although patients with a poor prognosis score had higher rates of palliative care use, under half received supportive care services, underscoring the need for efforts to prospectively target these patients with interventions that address their supportive care needs.

Patient-reported care satisfaction and symptom burden in hospitalized patients with cancer

180

Background: Hospitalized patients with cancer often experience a high symptom burden, which may impact care satisfaction and healthcare utilization. However, research describing these patients’ care satisfaction, symptom burden, and health care use is lacking. We sought to investigate relationships among care satisfaction, physical and psychological symptom burden, and hospital length of stay (LOS) in hospitalized patients with cancer. Methods: We prospectively enrolled patients with cancer and unplanned hospitalizations from 9/2014 to 4/2017. Upon admission, we assessed patients’ care satisfaction (FAMCARE items: satisfaction with care coordination and the speed with which symptoms are treated) as well as their physical (Edmonton Symptom Assessment System [ESAS]) and psychological (Patient Health Questionnaire-4 [PHQ-4]) symptoms. We used regression models to identify factors associated with care satisfaction, and we also examined associations of care satisfaction with patients’ symptom burden and hospital length of stay (LOS). Results: Among 1,576 participants (median age = 65.0 years [range:19-96], 46.3% female, 70.9% with incurable cancer, 58.4% admitted to a dedicated oncology service), most reported being “satisfied” or “very satisfied” with care coordination (90.1%) and the speed with which symptoms are treated (89.0%). Older age (care coordination: B < 0.01, P = 0.022, speed with which symptoms are treated: B = 0.01, P = 0.001) and admission to a dedicated oncology service (B = 0.20, P < 0.001 for each) were associated with higher care satisfaction. Higher satisfaction with care coordination was associated with lower ESAS-physical (B = -1.28, P = 0.007), ESAS-total (B = -2.73, P < 0.001), PHQ4-depression (B = -0.14, P = 0.022), and PHQ4-anxiety (B = -0.16, P = 0.008) symptoms. Higher satisfaction with the speed with which symptoms are treated was associated with lower ESAS-physical (B = -1.32, P = 0.003), ESAS-total (B = -2.46, P < 0.001), PHQ4-depression (B = -0.14, P = 0.014), and PHQ4-anxiety (B = -0.17, P = 0.004) symptoms. Greater satisfaction with care coordination (B = -0.48, P = 0.040) and the speed with which symptoms are treated (B = -0.44, P = 0.041) were both associated with shorter LOS. Conclusions: Hospitalized patients with cancer report high care satisfaction, which correlates with older age and admission to a dedicated oncology service. Significant associations among higher care satisfaction, lower symptom burden, and shorter hospital LOS highlight the importance of improving symptom management and care coordination in this population.

Supportive care services and goals of care in early phase clinical trials (EP-CTs)

26

Background: EP-CTs investigate novel therapeutic approaches for patients with cancer, but little is known about the use of supportive care services and timing of goals of care (GOC) discussions in EP-CTs. Methods: We conducted a retrospective review of consecutive patients with cancer enrolled on EP-CTs at Massachusetts General Hospital from 2017-2019. We collected information about patients’ demographic/clinical characteristics, use of supportive care services (palliative care [PC], social work [SW], physical therapy [PT], and nutrition), as well as documentation of GOC discussions and code status (before/during EP-CT vs after/never) via chart review. We examined patient characteristics associated with earlier receipt of supportive care services (before/during EP-CT vs after/never) and compared differences in the timing of GOC discussions and code status documented based on the receipt of supportive care services. Results: Among 425 patients enrolled on EP-CTs (median age 63.0; 56.0% female; 97.4% metastatic cancer; 22.1% gastrointestinal cancer), under half received supportive care services before/during trial (PC: 33.2% before/during, 66.8% post/never; SW: 41.9% before/during, 58.1% post/never; PT: 38.4% before/during, 61.6% post/never; and Nutrition: 33.2% before/during, 62.1% post/never). We identified the most common reasons for consulting each of the supportive care services (PC: 82.4% symptom management and 12.4% GOC; SW: 65.3% adjustment to illness and 23.8% referral for resources; PT: 44.8% safety/discharge planning and 24.6% mobility concerns; Nutrition: 73.2% for symptoms of anorexia/poor appetite and 21.5% nutrition assessment). Patients with GI cancer were more likely than those with other cancers to receive PC and SW before/during EP-CT (PC: 29.8% v 18.3%, p =.009; SW: 27.5% v 18.2%, p =.025). Earlier PC was associated with earlier hospice referral (HR = 1.95, p =.014) and shorter survival (HR = 1.54, p <.001). Patients receiving earlier supportive care services were more likely to have GOC discussions documented earlier (PC: 65.2% v 13.0%, p <.001; SW: 41.0% v 22.7%, p <.001; PT: 38.7% v 25.2%, p =.005; Nutrition: 39.1% v 25.0%, p =.002). Patients with earlier PC were more likely to have earlier documented code status (46.8% v 24.3%, p <.001), but not for any other service. Conclusions: In this cohort of patients with advanced cancer, under half received supportive care services before/during their participation in EP-CTs. We found that symptom management represented a common reason for referral to supportive care, highlighting the needs of this population. Patients who received earlier supportive care services were more likely to have earlier documentation of GOC discussions, with those receiving earlier PC having code status documented earlier and also experiencing earlier hospice use and shorter survival. These findings underscore the utility of supportive care services in EP-CTs.

Association of social support with overall survival and health care utilization in patients with aggressive hematologic malignancies

e18582

Background: Social support plays a crucial role for patients with aggressive hematologic malignancies as they navigate their illness course. We examined associations of social support with overall survival and health care utilization in this population. Methods: We conducted a cross sectional secondary analysis using data from a prospective longitudinal cohort study of 251 hospitalized patients with aggressive hematologic malignancies at Massachusetts General Hospital from 2014 through 2017. We utilized Natural Language Processing (NLP) to identify extent of patients’ social support (limited versus adequate as defined by NLP-aided chart review of the Electronic Health Record (EHR)). We used multivariable regression models to examine associations of social support with: 1) overall survival; 2) death or readmission within 90 days of discharge from index hospitalization; 3) time to readmission within 90 days; and 4) index hospitalization length of stay. Results: Patients had a median age of 64 (range: 19-93) years, and most were white (89.6%), male (68.9%), and married (65.3%). A plurality of patients had leukemia (42.2%) followed by lymphoma (37.9%) and myelodysplastic syndrome/myeloproliferative neoplasm (19.9%). Using NLP, we identified that 8.8% (22/251) of patients had limited social support. In multivariable analyses, limited social support was associated with worse overall survival (HR = 2.00, p = 0.042) and higher likelihood of death or readmission within 90 days of discharge (OR = 3.11, P = 0.043), but not with time to readmission within 90 days, or index hospital length of stay. Conclusions: In this cohort of hospitalized patients with aggressive hematologic malignancies, we found associations of limited social support with lower overall survival and higher likelihood of death or readmission within 90 days of hospital discharge. These findings underscore the utility of NLP for evaluating extent of social support and the need for larger studies evaluating social support in patients with aggressive hematologic malignancies.

Code status transitions in patients with high-risk acute myeloid leukemia (AML)

e19009

Background: Patients with high-risk AML often experience aggressive medical care at the end of life (EOL) such as hospitalization and intensive care unit (ICU) admission. Despite this, studies examining code status transitions in this population are lacking. Methods: We conducted a mixed methods study of 107 patients with high-risk AML enrolled in supportive care studies at Massachusetts General Hospital between 2014-2019. High-risk AML was defined as 1) new diagnosis > 60 years or 2) relapsed/refractory AML. Two physicians used consensus-driven medical record review to characterize code status transitions. Code statuses were coded as ‘full’ (confirmed or presumed), ‘restricted’ (i.e., do not resuscitate), or ‘comfort measures only’ (CMO); confirmations of presumed status were not coded as transitions. Results: At diagnosis of high-risk AML, 91.9% of patients were ‘full code’ (48.5% presumed, 43.4% confirmed) and 8.1% had restrictions on life-sustaining therapies. Overall, 55.1% (59/107) of patients experienced a code status transition, with a median of two transitions (range 1-4). Median time from first to last transition was 11 days (range 1-306) and from last transition to death was 1 day (range 0-11). Most of these transitions (79.6%; 48/59) were transitions to CMO near EOL. We identified three processes leading to code status transitions (Table): 1) pre-emptive conversations prior to any clinical change (15.3%; 9/59); 2) anticipatory conversations at the time of acute clinical deterioration (15%; 9/59); and 3) futility conversations after acute clinical deterioration, focused on withdrawing life-sustaining therapy (64.4%; 38/59). Only 55.9% (33/59) of patients participated in their last code status transition and 22.0% (13/59) of these transitions occurred in the ICU or Emergency Room. Conclusions: Most patients with high-risk AML had code status transitions at EOL, often following clinical deterioration that limited their ability to engage in EOL discussions. Interventions to promote earlier and more specific code status conversations are needed to improve patients’ ability to voice their EOL preferences.[Table: see text]

Changes in patient-reported outcomes (PROs) and tumor markers (TMs) to predict treatment response and survival outcomes in patients with metastatic gastrointestinal (GI) cancer

6560

Background: PROs assessing quality of life (QOL) and symptoms at a single timepoint frequently correlate with clinical outcomes in patients with cancer, yet efforts to understand how longitudinal changes in PROs can predict for treatment outcomes are lacking. In practice, oncologists often use changes in serum TMs (CEA and CA19-9) to monitor patients with GI cancer, and thus we sought to examine associations of 1-month changes in PROs and TMs with treatment response and survival outcomes among patients with advanced GI cancer. Methods: We prospectively enrolled patients with metastatic GI cancer prior to initiating chemotherapy at Massachusetts General Hospital from 5/2019-12/2020. At baseline (start of treatment) and 1-month later, we collected PROs (QOL [Functional Assessment of Cancer Therapy General {FACT-G}], physical symptoms [Edmonton Symptom Assessment System {ESAS}], and psychological symptoms [Patient Health Questionnaire-4 {PHQ-4}]) and TMs. We used regression models to examine associations of 1-month changes in PROs and TMs with treatment response (clinical benefit [defined as decreased or stable tumor burden] or progressive disease at the time of first scan) and survival outcomes (progression-free survival [PFS] and overall survival [OS]), adjusted for baseline values of each respective variable. Results: We enrolled 159 of 191 patients approached (83.2% enrollment); 134 had 1-month follow-up data (median age = 64 years [range: 28 to 84 years], 64.2% male, 46.3% pancreaticobiliary cancer). For treatment response, 63.4% had clinical benefit and 36.6% had progressive disease at the time of first scan (mean time to first scan = 2.01 months). Changes in PROs (ESAS-Total: OR = 0.97, p = 0.022; ESAS-Physical: OR = 0.96, p = 0.027; PHQ-4 depression: OR = 0.67, p = 0.014; FACT-G: OR = 1.07, p = 0.001), but not TMs (CEA: OR = 1.00, p = 0.836 and CA19-9: OR = 1.00, p = 0.796), were associated with clinical benefit at the time of first scan. Changes in ESAS-Total (HR = 1.03, p = 0.004), ESAS-Physical (HR = 1.03, p = 0.021), PHQ-4 depression (HR = 1.22, p = 0.042), FACT-G (HR = 0.97, p = 0.003), and CEA (HR = 1.00, p = 0.001) were predictors of PFS. Changes in ESAS-Total (HR = 1.03, p = 0.006) and ESAS-Physical (HR = 1.04, p = 0.015) were predictors of OS, but 1-month changes in TMs (CEA: HR = 1.00, p = 0.377 and CA19-9: HR = 1.00, p = 0.367) did not significantly predict for OS. Conclusions: We found that 1-month changes in PROs can predict for treatment response and survival outcomes in patients with advanced GI cancers. Notably, 1-month changes in CEA only correlated with PFS, while changes in CA19-9 did not significantly predict treatment response or survival outcomes. These findings highlight the potential for early changes in PROs to predict treatment outcomes while underscoring the need to monitor and address PROs in patients with advanced cancer.

Clinical and mutational profile of ARID1A-mutated gastrointestinal cancers: Duration of response to platinum-based chemotherapy

e15611

Background: ARID1A is mutated in several cancer types, with studies reporting mutations in up to 10% of colorectal cancers (CRC) and as high as 35% of gastric and pancreatic cancers. The ARID1A gene encodes a member of the SWI/SNF (SWItch/Sucrose Non-Fermentable) chromatin remodeling complex and functions as a tumor suppressor. ARID1A has also been implicated in double-stranded DNA repair via both homologous recombination and non-homologous end-joining, potentially conferring platinum sensitivity. We sought to characterize this subset of gastrointestinal (GI) malignancies. Methods: We identified patients with locally advanced or metastatic ARID1A-mutated GI malignancies treated at Massachusetts General Hospital (MGH) by next-generation sequencing. Patients were selected who gave consent to molecular testing and who were enrolled on to a study. We evaluated clinical characteristics and outcomes for patients undergoing treatment at MGH between 2009 and May 2020. The Kaplan-Meier method was used to calculate progression free survival (PFS) to first-line platinum-based chemotherapy. Results: We captured 38 patients with ARID1A-mutated tumors. Median age at diagnosis was 66 (range 31-87) and 63.2% of patients were male (n = 24). Tumor types varied, including CRC (n = 13, 34.2%), esophagogastric (n = 13, 34.2%), pancreatic (n = 6, 15.7%), cholangiocarcinoma (n = 2, 5.3%), small bowel (n = 1, 2.6%), anal (n = 1, 2.6%), and unknown GI primary (n = 2, 5.3%). Most were metastatic at diagnosis (n = 23, 60.5%). The identified ARID1A mutations were each distinct, occurring along the length of the gene and were comprised of missense (n = 10, 26.3%), nonsense (n = 12, 31.6%), frameshift (n = 13, 34.2%), and splice-site (n = 3, 7.9%) mutations. We observed on average 4-5 co-mutations per tumor, with TP53 (n = 25, 65.8%), KRAS (n = 14, 36.8%), APC (n = 11, 28.9%), BRCA2 (n = 7, 18.4%) and BRAF (n = 7, 18.4%) occurring most frequently. Tumors were both microsatellite stable (n = 23, 60%) and microsatellite unstable (n = 7, 18.4%). Most patients (n = 37, 97.4%) received a platinum-based chemotherapy as first-line therapy including FOLFOX (n = 23, 60.5%), FOLFIRINOX (n = 10, 26.3%), gemcitabine/cisplatin (n = 2, 5.3%), carboplatin/5-FU (n = 1, 2.6%), and carboplatin/etoposide (n = 1, 2.6%). Median PFS for first-line platinum based chemotherapy was 14.0 months (CI 8.2-34.7) overall. For patients with CRC, PFS to platinum-based therapy was 14.0 months (CI 4.8-not reached) compared with 9.6 months for non-CRC (CI 7.4-not reached). Conclusions: To our knowledge, this is the first assessment of clinical characteristics and outcomes for ARID1A-mutated GI malignancies. Mutations in ARID1A are highly diverse, without a clear association with tumor type. Future studies assessing response to platinum-based chemotherapy are warranted.

Financial toxicity, symptom burden, illness perceptions, and communication confidence in cancer clinical trial participants

6526

Background: Cancer clinical trial (CCT) participants are at high risk for experiencing adverse effects from financial toxicity, yet this remains understudied in the CCT population. We sought to describe associations among patient-reported financial toxicity (financial burden [FB] and trial cost concerns), physical and psychological symptoms, illness perceptions, and communication confidence in CCT participants. Methods: From 7/2015-7/2017, we prospectively enrolled CCT participants who expressed interest in financial assistance and a group of patients matched by age, sex, cancer type, specific trial, and trial phase. We assessed FB (burdened by costs of cancer care), trial cost concerns (worried about affording medical costs of a CCT), physical (Edmonton Symptom Assessment Scale [ESAS]) and psychological (Patient Health Questionnaire-4 [PHQ-4]) symptoms, illness perceptions (Brief Illness Perception Questionnaire [BIPQ]), and communication confidence (Perceived Efficacy in Patient-Physician Interactions [PEPPI]). We used regression models to explore sociodemographic associations with FB and trial cost concerns, and to examine associations of FB and trial cost concerns with patients’ symptom burden, illness perceptions, and communication confidence, adjusting for age, sex, race, performance status, marital status, and metastatic status. Results: Of 198 patients enrolled, 112 (56.6%) reported FB and 82 (41.4%) had trial cost concerns. Patients with FB were younger (OR 0.96, 95% CI 0.94-0.98) and had lower incomes ( < $100,000, OR 4.61, 95% CI 2.35-9.01). Patients reporting trial cost concerns also had lower incomes ( < $100,000, OR 2.78, 95% CI 1.45-5.29). On adjusted analyses, patients with FB had higher ESAS total (OR 1.03, 95% CI 1.02-1.05), ESAS physical (OR 1.04, 95% CI 1.02-1.07), PHQ-4 depression (OR 1.54, 95% CI 1.22-1.94), and PHQ-4 anxiety (OR 1.30, 95% CI 1.08-1.55) scores, as well as more negative illness perceptions (OR 1.04, 95% CI 1.01-1.07), but no significant difference in communication confidence (OR 0.98, 95% CI 0.93-1.05). Patients reporting trial cost concerns had higher ESAS total (OR 1.03, 95% CI 1.01-1.05), ESAS physical (OR 1.04, 95% CI 1.01-1.06), PHQ-4 depression (OR 1.35, 95% CI 1.10-1.65), and PHQ-4 anxiety (OR 1.27, 95% CI 1.07-1.51) scores, as well as more negative illness perceptions (OR 1.06, 95% CI 1.03-1.10), and lower communication confidence (OR 0.93, 95% CI 0.87-0.99). Conclusions: In this study of CCT participants, younger patients with lower incomes were most vulnerable to financial toxicity. Financial toxicity was associated with greater symptom burden, more negative illness perceptions, and lower communication confidence, which underscores the importance of addressing these issues when seeking to alleviate the adverse effects of financial toxicity in CCT participants.

Survival outcomes, treatment toxicity, and healthcare utilization in older adults with aggressive non-Hodgkin lymphoma (NHL)

7557

Background: Aggressive NHLs frequently affect older adults, and are often treated with intensive systemic therapy that is potentially curative but can cause substantial toxicities. Although balancing treatment efficacy with the risk of complications is critically important for older adults with NHL, few studies have described these patients’ survival outcomes, rates of toxicities, and healthcare utilization. Methods: We conducted a retrospective analysis of adults > 65 years diagnosed with aggressive NHL and treated with systemic therapy at Massachusetts General Hospital from 4/2000-7/2020. We abstracted patient demographic and clinical information, survival outcomes, treatment toxicity (rates and grade), and healthcare utilization outcomes (intensive care unit [ICU] admissions and unplanned hospitalizations within six months of treatment initiation) from the electronic health record. Using multivariable logistic regression, we examined patient and disease factors associated with rates of grade 3+ non-hematologic toxicity and unplanned hospitalization. Results: Of 295 patients (median age = 73 years [age 65-69: 32.5%; age 70-74: 26.1%; age 75-79: 20.0%; age 80+: 21.4%], 39.0% female), most had advanced stage disease (59.5%) and an ECOG performance status of 0 or 1 (83.1%). The most common diagnosis was de novo diffuse large B-cell lymphoma (DLBCL) or grade 3B follicular lymphoma (69.2%). Most common therapies were CHOP (65.8%) and EPOCH (17.0%) with or without Rituximab. With a median follow up of 5.9 years, 5-year overall survival (OS) was 74.2%. Among patients age 65-69, 70-74, 75-79, and 80+ years, 5-year OS by age group were 82.1%, 72.2%, 73.5%, and 66.3%, respectively. Overall, 42.4% had grade 3+ toxicity, while 8.1% had grade 4 or 5 toxicity. The rates of unplanned hospitalization and ICU admission during the first 6 months of therapy were 41.0% and 6.1%, respectively. In multivariable analysis, hypoalbuminemia (OR 4.22, 95%, p < 0.001) and number of comorbidities (OR 1.75, p < 0.001) were associated with a greater likelihood of grade 3+ toxicity. Hypoalbuminemia (OR 2.76, p = 0.003), number of comorbidities (OR 1.61, p = 0.001), and receipt of EPOCH (OR 5.41, p = 0.012) were associated with a greater likelihood of unplanned hospitalization. Conclusions: The majority of older adults receiving upfront therapy for aggressive NHL survive beyond 5 years, yet nearly half experience substantial treatment toxicities and unplanned hospitalizations. Our findings underscore the need to develop supportive care interventions to enhance the care experience for older adults with NHL.

Association between perceptions of prognosis and end-of-life outcomes for patients with advanced lung and gastrointestinal cancer

6503

Background: Many patients with advanced cancer maintain misperceptions of their prognosis and are thus unprepared to make difficult decisions regarding their end-of-life (EOL) care. However, studies examining the associations between patients’ perceptions of their prognosis and their EOL outcomes are limited. Methods: We conducted a secondary analysis using longitudinal data from a randomized controlled trial of a palliative care intervention for patients with newly diagnosed incurable lung and non-colorectal gastrointestinal cancer. We administered the Prognosis and Treatment Perceptions Questionnaire to assess patients’ perceptions of their prognosis at baseline, week-12, and week-24, using the final assessment closest to death. We used multivariate logistic and linear regression models, adjusting for age, gender, marital status, cancer type, and randomization to the palliative care intervention, to examine the associations among patients’ perceptions of their prognosis with the following EOL care outcomes abstracted from the electronic health record: 1) hospice utilization and length-of-stay (LOS); 2) hospitalizations in the last 30 days of life; 3) receipt of chemotherapy in the last 30 days of life; and 4) location of death. Results: We enrolled 350 patients in the parent trial, of which 80.5% (281/350) died during the study period and were included in this analysis. Overall, 59.4% (164/276) of patients reported that they were terminally ill, and 66.1% (154/233) reported that their cancer was likely curable at the assessment closest to death. In multivariate analyses, patients who reported that their cancer was likely curable were less likely to utilize hospice (OR = 0.25, 95%CI 0.10-0.61, P = 0.002) or die at home (OR = 0.56, 95%CI 0.32-0.98, P = 0.043), and more likely to be hospitalized in the last 30 days of life (OR = 2.28, 95%CI 1.20-4.32, P = 0.011). In contrast, patients’ report that they were terminally ill was only associated with lower likelihood of hospitalizations in the last 30 days of life (OR = 0.52, 95%CI 0.29-0.92, P = 0.025). Patients’ perceptions of their prognosis were not associated with hospice LOS or chemotherapy administration in the last 30 days of life. Conclusions: Patients’ perceptions of their prognosis are associated with important EOL outcomes including hospice utilization, hospitalizations at the EOL, and death at home. Interventions are needed to enhance patients’ perceptions of their prognosis in order to optimize their EOL care.

Local and systemic recurrence following total neoadjuvant therapy (TNT) and resection for borderline resectable and locally advanced pancreatic adenocarcinoma: Long-term follow up from two phase II studies

4133

Background: With the advent of FOLFIRINOX, the management of pancreatic cancer has undergone a profound change. There has been a shift to TNT with FOLFIRINOX followed by radiation and an attempt at surgical resection. Recent trials of TNT have demonstrated an ability to resect locally advanced (LA) and borderline resectable disease. There is a lack of prospective data demonstrating local and systemic recurrence rates after TNT. Methods: Two previously reported prospective clinical trials (Murphy JE, et al, JAMA Oncol 2018, 2019) of total neoadjuvant therapy were conducted between 2012 and 2018 for borderline and LA disease (NCT01591733, NCT01821729). Patients received FOLFIRINOX for 8 cycles. Upon restaging, patients with resolution of vascular involvement received short-course chemoradiotherapy (5 Gy x 5 with protons or 3 Gy x 10 w photons) with capecitabine (N=34). Patients with persistent vascular involvement received long-course chemoradiotherapy with capecitabine (N=56). All patients were considered for resection after TNT except for those patients with metastatic or unresectable disease. Results: 97 eligible patients were enrolled and started treatment on the borderline resectable (n = 48) and locally advanced (n= 49) study. 90 patients completed therapy. 80 patients were taken to the operating room. 61 patients had R0 resection and 5 patients had R1 resection. The table shows the distribution of local recurrences, local recurrences and metastatic disease, and metastatic disease alone. With a median follow-up of 5.2 years (range: 2.4-6.0), of the 61 R0 patients, 22 patients remained alive and free of disease, 7 patients had a local recurrence, 4 patients had locoregional and metastatic recurrence, and 24 patients had a metastatic recurrence. 3 patients who underwent R0 resection died of unrelated causes. Median survival for patients undergoing R0 resection is 43.8 months. Conclusions: Total neoadjuvant therapy for locally advanced and borderline resectable pancreatic cancer is potentially curable and may change the pattern of spread.[Table: see text]

Supportive oncology care at home intervention for patients with pancreatic cancer

6558

Background: Patients with pancreatic cancer receiving chemotherapy often experience substantial symptoms and high healthcare utilization. We sought to determine the feasibility of delivering a Supportive Oncology Care at Home intervention designed to address the needs of patients receiving treatment for pancreatic cancer. Methods: We prospectively enrolled patients with pancreatic cancer who were participating in a parent trial of neoadjuvant FOLFIRINOX and residing in-state, within 50 miles of our hospital. Patients received the Supportive Oncology Care at Home intervention during neoadjuvant treatment (i.e., up to 4 months). The intervention entailed: 1) remote monitoring of daily patient-reported symptoms, daily vital signs, and weekly body weight; 2) a hospital in the home care model for symptom assessment and management; and 3) structured communication with the oncology team. We defined the intervention as feasible if ≥60% of patients enrolled in the study and ≥60% completed the daily assessments within the first two weeks of enrollment. We tracked numbers of phone calls, emails, and home visits generated by the intervention. We conducted exit interviews with patients, caregivers, and oncology clinicians to assess the acceptability of the intervention. In addition, we compared rates of treatment delays, urgent clinic visits, emergency room (ER) visits, and hospitalizations among those who did (n = 20) and did not (n = 24) receive Supportive Oncology Care at Home from the parent trial. Results: From 1/2019-9/2020, we enrolled 80.8% (21/26) of potentially eligible patients. One patient became ineligible following consent due to moving out-of-state, resulting in 20 participants (median age = 67 years [range 55-77]; 60.0% female). Within the first two weeks of enrollment, 65.0% completed all the daily assessments, with participants reporting 96.1% of daily symptoms, 96.1% of daily vital signs, and 92.5% of weekly body weights. Each participant generated an average of 2.22 phone calls (range 0.62-3.77), 2.96 emails (range 1.50-5.88), and 0.15 home visits (range 0-0.69) per week. During exit interviews, > 80% of patients, caregivers, and clinicians found the intervention to be helpful and convenient, and they reported high satisfaction with the communication among patients, clinicians, and the hospital in the home team. Patients receiving the intervention had lower rates of treatment delays (55.0% v 75.0%), urgent clinic visits (10.0% v 25.0%), ER visits or hospitalizations (45.0% v 62.5%), as well as a lower proportion of days spent in urgent clinic, ER, or hospital (2.7% v 7.8%), compared with those not receiving the intervention who were in the same parent trial. Conclusions: These findings demonstrate the feasibility and acceptability of a Supportive Oncology Care at Home intervention. Future work will investigate the efficacy of this intervention for decreasing healthcare use and improving patient outcomes. Clinical trial information: NCT03798769.

Association between baseline geriatric domains and survival in older adults with chronic lymphocytic leukemia (CLL)

12041

Background: CLL is a disease that commonly affects older adults. Although the value of geriatric assessment is increasingly being recognized in older adults with cancer, few studies have examined the relationship between baseline geriatric domains and clinical outcomes in older adults with CLL. Methods: We conducted a secondary data analysis of 369 adults diagnosed with CLL and treated in a phase 3 randomized trial of patients age ≥65 with bendamustine plus rituximab versus ibrutinib plus rituximab versus ibrutinib alone. We evaluated geriatric domains of functional status (activities of daily living [ADL], instrumental activities of daily living [IADL], Timed “Up and Go,” and number of falls in last 6 months), psychological status (Mental Health Inventory), social activity (Medical Outcomes Study [MOS] Social Activity Survey), cognition (Blessed Orientation Memory Concentration Test), social support (MOS Social Support Tangible and Emotional/Informational subscales), and nutritional status ( > 5% weight loss in the preceding 6 months). We examined associations among baseline geriatric domains with overall survival (OS) and progression-free survival (PFS) using multivariable Cox regression models. Results: The median age of patients was 71 years (range: 65-89). Most were male (67.1%) and had an ECOG performance status of 0 or 1 (96.9%). In multivariable models, the following geriatric domains were significantly associated with OS: better functional status (ADL score: HR 0.67, p = 0.012; IADL score: HR 0.98, p = 0.007); social activity score (HR 0.97, p = 0.004); and nutritional status (HR 2.58, p = 0.008). Similarly, functional status (ADL score: HR 0.77, p = 0.028; IADL score: HR 0.99, p = 0.007); social activity score (HR 0.97, p < 0.001); and nutritional status (HR 2.87, p < 0.001) were all associated with PFS. Additionally, the number of impaired geriatric domains was also associated with OS (HR 1.50, p = 0.004) and PFS (HR 1.45, p < 0.001). Timed “Up and Go”, number of falls in last 6 months, psychological status, cognition, and social support were not significantly associated with clinical outcomes. Conclusions: Geriatric domains of functional status, social activity, and nutritional status were associated with OS and PFS in this cohort of older adults with CLL. These findings highlight the importance of assessing geriatric domains to identify high-risk patients with CLL who may benefit from additional support during their treatment.

Symptom burden, functional status, and clinical outcomes of hospitalized patients with advanced genitourinary cancers

42

Background: Patients with advanced genitourinary (GU) cancers are often hospitalized for complications of their cancer and symptom management. Yet, little is known about the symptom burden, functional status, and health care utilization of these patients. Methods: We prospectively enrolled patients with advanced cancer who experienced unplanned hospitalizations at an academic medical center. Upon admission, we asked patients to self-report their physical (Edmonton Symptom Assessment Scale-revised [ESAS-r]) and psychological (Patient Health Questionnaire-4 [PHQ-4]) symptoms. We also collected data from nursing assessments about impairments in activities of daily living (ADLs). We compared symptoms, functional impairment, readmissions, and overall survival (OS) between cancer types (dichotomizing GU cancers vs other cancer types) and within GU cancers (dichotomizing prostate cancer vs kidney/bladder/adrenal cancer) using univariate and multivariable regression analyses adjusted for age, sex, education, comorbidities, and time since advanced cancer diagnosis. Results: Among 971 patients enrolled, 106 (10.9%) had advanced GU cancers (39.6% prostate cancer, 32.1% kidney cancer, 25.5% bladder cancer, and 2.8% adrenal cancer). Compared to patients with other cancer types, patients with GU cancers were older (median: 69.0 vs 64.0 years, p < 0.001) and had more time since advanced cancer diagnosis (median: 14.0 vs 7.0 months, p < 0.001). In univariate analyses, a greater proportion of patients with GU cancers had an ADL impairment (57.5% vs 38.0%, p < 0.001) compared to other cancer types but the groups did not differ in their physical (Mean = 33.3 vs 32.6, p = 0.61) or depression (Mean = 4.1 vs 3.3, p = 0.05) symptoms. In multivariable models, patients with GU cancers had similar risk of readmission in 90 days (HR 1.31, p = 0.077), but worse survival (median OS: 102.0 days vs 133.5 days, p < 0.001; HR 1.27, p = 0.046). Within GU cancers, patients with kidney/bladder/adrenal cancer (vs. prostate cancer) were younger (median: 66.0 vs 74.0, p < 0.001) with less time since advanced cancer diagnosis (median: 9.0 vs 23.0 months, p = 0.012) but had no difference in symptoms or functional impairment. They were more likely to be admitted for symptom management (66% vs. 39% for prostate cancer, p = 0.026). Patients with kidney/bladder/adrenal cancer also had higher risk of readmission (HR 2.04, p = 0.043) but no difference in OS, compared to patients with prostate cancer. Conclusions: We found that hospitalized patients with advanced GU cancers had significantly greater functional impairment and worse survival compared to those with other cancer types, and those with kidney/bladder/adrenal cancer had significantly higher readmission risk compared to those with prostate cancer. These findings support the need to develop tailored supportive care for hospitalized patients with GU cancers.

Differential effects of an electronic symptom monitoring intervention based on the age of patients with advanced cancer

BACKGROUND

Symptom monitoring interventions enhance patient outcomes, including quality of life (QoL), health care utilization, and survival, but it remains unclear whether older and younger patients with cancer derive similar benefits. We explored whether age moderates the improved outcomes seen with an outpatient electronic symptom monitoring intervention.

PATIENTS AND METHODS

We carried out a secondary analysis of data from a randomized trial of 766 patients receiving chemotherapy for metastatic solid tumors. Patients received an electronic symptom monitoring intervention integrated with oncology care or usual oncology care alone. The intervention consisted of patients reporting their symptoms, which were provided to their physicians at clinic visits, and nurses receiving alerts for severe/worsening symptoms. We used regression models to determine whether age (older or younger than 70 years) moderated the effects of the intervention on QoL (EuroQol EQ-5D), emergency room (ER) visits, hospitalizations, and survival outcomes.

RESULTS

Enrollment rates for younger (589/777 = 75.8%) and older (177/230 = 77.0%) patients did not differ. Older patients (median age = 75 years, range 70-91 years) were more likely to have an education level of high school or less (26.6% versus 20.9%, P = 0.029) and to be computer inexperienced (50.3% versus 23.4%, P < 0.001) compared with younger patients (median age = 58 years, range 26-69 years). Younger patients receiving the symptom monitoring intervention experienced lower risk of ER visits [hazard ratio (HR) = 0.74, P = 0.011] and improved survival (HR = 0.76, P = 0.011) compared with younger patients receiving usual care. However, older patients did not experience significantly lower risk of ER visits (HR = 0.90, P = 0.613) or improved survival (HR = 1.06, P = 0.753) with the intervention. We found no moderation effects based on age for QoL and risk of hospitalizations.

CONCLUSIONS

Among patients with advanced cancer, age moderated the effects of an electronic symptom monitoring intervention on the risk of ER visits and survival, but not QoL. Symptom monitoring interventions may need to be tailored to the unique needs of older adults with cancer.

Patient-reported outcomes (PROs) in older adults with gastrointestinal (GI) cancer undergoing surgery

159

Background: Older adults with GI cancer often experience poor surgical outcomes, yet little is known about their PROs, such as physical function, quality of life (QOL), and physical and psychological symptom burden. Methods: As part of a randomized trial of perioperative geriatric care, we prospectively enrolled adults age ≥65 with GI cancer planning to undergo surgical resection. We asked patients preoperatively to self-report their physical function (activities of daily living [ADLs] and instrumental ADLs [IADLs]), QOL (EORTC QLQ-C30), symptom burden (Edmonton Symptom Assessment System [ESAS], scores > 3 considered moderate/severe [mod/sev]), depression symptoms (Geriatric Depression Scale [GDS], scores > 4 represent a positive screen for depression), and comorbidities. We used regression models to explore relationships among PROs and clinical outcomes (receiving planned surgery, postoperative complications [Clavien-Dindo], hospital readmissions within 90 days, and survival). Results: From 9/2016 - 4/2019, we enrolled 160 of 221 (72.4%) patients approached (median age: 72, range: 65-92). At baseline, most (53.1%) reported at least one comorbidity and required assistance with ADLs (94.8%) and IADLs (52.3%). Patients reported an average of 2.56 mod/sev ESAS symptoms, and 27.7% screened positive for depression. For surgical outcomes, 137 patients (85.6%) underwent planned surgery, and 99 (72.2%) of these had at least one postoperative complication. Greater independence with ADLs was associated with undergoing planned surgery (OR = 1.21, P = .02), lower risk of complications (OR = 0.81, P < .01), and improved survival (HR = 0.87, P = .02), but not readmissions. Greater independence with IADLs was associated with undergoing planned surgery (OR = 1.30, P = .03) and improved survival (HR = 0.73, P < .01), but not other outcomes. Higher baseline QOL was only associated with lower risk of postoperative complications (OR = 0.97, P = .04). Higher depression scores were only associated with worse survival (HR = 1.13, P = .02). Higher baseline symptom burden predicted for shorter time to readmission (HR = 1.13, p = .03). Patient-reported number of comorbidities was associated with shorter time to readmission (HR = 1.49, p = .03) and higher risk of complications (OR = 1.70, P = .03). Conclusions: Older adults with GI cancer often have baseline functional limitations and a high symptom burden, all of which are associated with worse clinical outcomes. Future work should study whether addressing preoperative PROs could improve older patients’ surgical outcomes. Clinical trial information: NCT02810652 .

Use of patient-reported outcomes (PROs) to predict treatment outcomes in patients with advanced cancer

186

Background: PROs assessing quality of life (QOL) and physical symptoms often correlate with clinical outcomes in patients (pts) with cancer. Yet, data are lacking about the use of PROs to predict treatment response. We evaluated associations of baseline PROs with treatment response, healthcare use, and survival among pts with advanced gastrointestinal cancer. Methods: We prospectively enrolled pts with metastatic gastrointestinal cancer prior to initiating chemotherapy at Massachusetts General Hospital. At baseline (start of treatment), pts reported their QOL (Functional Assessment of Cancer Therapy General [FACT-G], subscales assess QOL across 4 domains: functional, physical, emotional, social well-being) and symptom burden (Edmonton Symptom Assessment System [ESAS]). Higher scores on FACT-G indicate better QOL, while higher scores on ESAS represent a greater symptom burden. We used regression models to examine associations of baseline PRO scores with treatment response (clinical benefit [CB] or progressive disease [PD] at the time of first scan based on clinical documentation), healthcare use (unplanned hospital admissions), and survival. Results: From 5/2019-3/2020, we enrolled 112 of 131 (85.5% enrollment) consecutive pts (median age = 62.8, 61.6% male, 45.5% pancreatobiliary cancer). For treatment response, 64.3% had CB and 35.7% had PD. Higher ESAS-physical (B = 1.04, p = .027) and lower FACT-G functional (B = 0.92, p = .038) scores at baseline were significant predictors of PD. On the specific ESAS items, pts who experienced PD were more likely to report moderate/severe poor well-being (57.9% vs 29.7%; p = .001), pain (44.7% vs 25.0%; p < .050), drowsiness (42.1% vs 20.3%; p = .024), and diarrhea (23.7% vs 4.7%; p = .008) at baseline. Lower FACT-G total (HR = 0.96, p = .003), FACT-G physical (HR = 0.89, p < .001), FACT-G functional (HR = 0.87, p < .001), and higher ESAS-physical (HR = 1.03, p = .028) scores at baseline were significantly associated with greater risk of hospital admission. Lower FACT-G total (HR = 0.96, p = .009), FACT-G emotional (HR = 0.87, p = .014), as well as higher ESAS-total (HR = 1.03, p = .018) and ESAS-physical (HR = 1.03, p = .040) scores at baseline were significantly associated with greater risk of death. Conclusions: We found that baseline PROs predict treatment response in pts with advanced cancer, namely physical symptoms and functional QOL, in addition to healthcare use and survival outcomes. These findings further support the use of PROs to predict important clinical outcomes, including the novel finding of treatment response.

Leveraging patient-reported outcomes (PROs) in patients with pancreatic cancer: The Pancreatic Cancer Action Network (PanCAN) online patient registry experience

154

Background: By allowing patients (pts) to self-report key issues related to their quality of life and symptoms, PROs have important clinical and research implications. The PanCAN Registry, which began collecting data in July 2015, is a pancreatic cancer-specific global online registry enabling pts to report sociodemographics, disease/management characteristics, and PROs via online surveys. We sought to describe pt experiences with the PanCAN Registry. Methods: We assessed individual characteristics and interactions with the registry (visits, survey completions, and longitudinal use) from 7/2015-10/2019 for pts who provided permission to use their data. The registry allows pts to complete surveys about their experience (e.g. basics of pancreatic cancer, general information), symptoms (e.g. fatigue, pain), diagnostics (e.g. labs, scans), and drug therapy (e.g. type, frequency). We validated PROs using the PanCAN Know Your Tumor database. For a subset of pts (those with de novo metastatic disease), we compared PROs, treatment patterns, and side effects by age (+/- 65 years) and treatment site (community or academic). Results: Of 2,836 pts who visited the registry, 2,076 (73%) completed at least one survey (median age = 64 [range: 18-97], 48% women, 92% white, 32% metastatic disease). Pts most commonly completed the basics (73%), general information (39%), and drug therapy (37%) surveys. Overall, 10% answered surveys longitudinally. We observed 95% concordance between PROs and the PanCAN Know Your Tumor database. Among the 667 pts with de novo metastatic disease, 34% were older (age 65+) and 50% were treated at academic sites. Younger pts were more hopeful about the treatment plan (strongly agree: 24% v 12%, p < .01) and reported less constipation (moderate/severe: 33% v 48%, p < .01) compared with older pts. Pts treated at academic sites reported less frequent treatment breaks of > 2 weeks (28% v 58%, p = 0.01) and more frequent severe cytopenias (27% v 12%, p = 0.01) compared with those treated at community sites. Conclusions: With > 2,800 pts visiting the PanCAN registry and > 70% completing a survey, these findings demonstrate the feasibility, robustness, and research potential of an online PRO registry. We observed important differences by age and treatment site regarding pts’ outlook, symptoms, treatment patterns, and side effects. With increasing focus on PROs, registries like this can facilitate standardized PRO reporting and monitoring, while also providing a valuable research database.

Caregiver-oncologist concordance in patient prognosis, caregiver depression, and caregiver mastery

143

Background: Caregivers of older adults with advanced cancer often have a different understanding of the patient’s prognosis compared with their oncologist. Among patients, accurate prognostic awareness is associated with greater depressive symptoms, except when patients utilize more adaptive coping skills. We examined the relationship between caregiver-oncologist prognostic concordance and caregiver depressive symptoms and explored whether this relationship differed by caregiver mastery, the capacity to cope, adjust, and adapt to problems. Methods: We utilized data from a national geriatric assessment cluster-randomized trial (URCC 13070: PI Mohile) that recruited patients aged ≥70 with incurable cancer considering any line of cancer treatment at community oncology practices, their caregivers, and oncologists. At enrollment, caregivers and oncologists estimated the patient’s prognosis (0-6 months, 7-12 months, 1-2 years, 2-5 years, > 5 years); same response was considered concordant. Caregivers completed Ryff’s mastery subscale (range 7-35, higher is better) at enrollment and depression screen (the Patient Health Questionnaire (PHQ)-2 (range 0-6) 4-6 weeks later. To assess the association of prognostic concordance with caregiver depressive symptoms, we used generalized estimating equations in models adjusted for cancer type, study arm, practice sites, and caregiver demographics. We then assessed moderation effect of caregiver mastery on this association. Results: Among 410 caregiver-oncologist dyads, mean caregiver age was 66.5, 75% were female, and 26% were caregivers of patients with lung cancer. Mean mastery score at enrollment was 27.6 (SD 4.7) and 19% screened positive on PHQ-2 at week 4-6. Among dyads who provided response (N = 370), 28% were concordant. Prognostic concordance was associated with higher caregiver depressive symptoms (β = 0.30; p = 0.04). Significant moderation effect was found between concordance and mastery for caregiver depressive symptoms (p = 0.02). Among caregivers with low mastery ( < median), prognostic concordance was associated with higher depressive symptoms (β = 0.68; p = 0.003). Among caregivers with high mastery (≥median), concordance was not associated with depressive symptoms (β = -0.06; p = 0.67). Conclusions: There is a need to study how prognostic understanding might lead to depression in at-risk caregivers. Interventions targeting caregiver prognostic understanding need to consider its relationship with depressive symptoms, while seeking to increase caregiver mastery.

Palliative care referrals in patients with advanced cancer on early-phase cancer clinical trials (EP-CTs)

29

Background: EP-CTs investigate novel therapeutic approaches for patients with cancer, but little is known about the utilization of supportive care services, specifically palliative care (PC), in this population. Methods: We conducted a retrospective review of consecutive patients enrolled on EP-CTs at the MGH Cancer Center from 2017-2019. Sociodemographic and clinical variables, including utilization of PC services, were obtained via chart review. Details of the PC evaluation were compared between patients who received first referral to PC while enrolled on an EP-CT versus those who received a PC referral at any point after diagnosis. Results: Among 426 patients enrolled on EP-CTs (median age 63 years; 44% male), 249 (59%) received a PC referral at any time following a diagnosis of cancer (median age 57 years, 58% male). Eighty-six (35%) were referred prior to enrollment on EP-CT, 44 (18%) were referred while on EP-CT, and 119 (48%) were referred post-EP-CT. Patients referred on EP-CT were younger (median 56 vs 63 years, p < 0.0001) than those enrolled on EP-CTs. For patients referred while on EP-CT, 48% had a PC consult within 30 days of enrollment (range: 0-530 days); median number of PC visits was 3 (range: 0-37); median time from first PC consult to death or termination of EP-CT was 32 days (range: 1-213). Of 44 patients referred on EP-CT, 2 (5%) died while on EP-CT. Of the remaining patients, median time from first PC consult to date of death was 79 days (IQR: 45-178 days). Most common reasons for referral included pain (22, 50%), non-pain symptoms (21, 48%), and goals of care/advanced care planning (20, 45%). Of these referrals, 13 (30%) were initiated as inpatients versus 31 (70%) as outpatients. Pain was most commonly cited for outpatient referral (35%), followed by non-pain symptoms (25%) and goals of care (23%). Non-pain symptoms (40%) and goals of care (36%) were most commonly cited reasons for inpatient referral, followed by pain (24%). Of referrals while on EP-CT, 23 (52%) were made by EP-CT staff, including MD and APP, 7 (16%) from the primary oncologist, and 1 (2%) was self-referral. 26 (57%) of patients referred to PC during trial were also referred to hospice, with a median time from last PC consult to hospice referral of 24 days (range: -2-322). Conclusions: A majority of patients with advanced cancer enrolled on EP-CTs received a PC referral. The timeline and method of referral varied, but most patients did not receive a referral until or following enrollment on an EP-CT. Future work will focus on developing a standard referral protocol for patients enrolled on EP-CTs.

Communicating the components of informed treatment decision-making in patients with pancreatic cancer receiving preoperative therapy

147

Background: Preoperative therapy for localized pancreatic cancer represents an emerging treatment paradigm with the potential to provide significant benefits, yet with complex risks. Research is lacking about whether clinicians effectively communicate key components of informed decision-making for patients considering this treatment. Methods: From 2017-2019, we conducted a two-part, mixed methods study. In part 1, we conducted interviews with clinicians (medical/radiation/surgical oncology, n = 13) and patients with pancreatic cancer who had received preoperative therapy (n = 18) to explore perceptions of information needed to make informed decisions about preoperative therapy, from which we generated a list of key elements. In part 2, we audio recorded the initial multidisciplinary visits of patients with pancreatic cancer eligible for preoperative therapy (n = 20). Two coders (94% concordance) independently identified whether clinicians discussed key elements from part 1. Patients also completed a post-visit survey reporting whether clinicians discussed the key elements. We explored discordance between audio recordings and patient reports using qualitative, explanatory themes. Results: In part 1, we identified 13 key elements of informed treatment decision-making, including treatment logistics, alternatives, and potential risks/benefits. In part 2, recordings showed that most visits included discussions about logistics, such as the chemotherapy schedule (n = 20) and use of a port-a-cath (n = 20), whereas few included discussions about risks, such as the potential for hospitalizations (n = 7), urgent visits (n = 6), or needing help with daily tasks (n = 6). Patients reported hearing about potential benefits, such as likelihood of achieving surgery (n = 10) and cure (n = 7), even when these were not discussed. Qualitative themes across these discordant cases included clinician optimism regarding present day results versus historical findings and mentions of positive outcomes from prior patients without citing specific data or potential adverse outcomes. Conclusions: We identified key elements of information patients with pancreatic cancer need to make informed decisions about preoperative therapy. Although clinicians frequently disclosed much of this information, we found multiple cases of patient-clinician discordance for certain key elements, which underscores the need for interventions to enhance patient-clinician communication regarding pancreatic cancer treatment decisions.

Involvement of social work services in patients with advanced cancer in early-phase clinical trials (EP-CTs)

28

Background: Early integration of supportive care in patients with advanced cancer has improved quality of life, symptom burden, and survival. Participants in EP-CTs often are highly pre-treated and the demands of participation can exacerbate financial and psychosocial concerns. Integration of social work services can address a broad scope of concerns including behavioral health, psychosocial needs, physical health, and financial concerns. Little data exists regarding the use of social work services in this patient population. Methods: We conducted a retrospective chart review of consecutive patients enrolled in EP-CTs at the MGH Cancer Center during 2017-2019. Information including sociodemographic data, clinical variables, and the use of social work services were captured from the electronic health record. We reviewed documentation in social work notes to determine reason for referral while participating on trial. Results: Of 426 EP-CT participants, 64% ( n = 272) received social work consultations at any time during their cancer course (consultations occurred a median of 23 months [range 0 – 444] following diagnosis). Compared to those who did not receive consultation, patients receiving consultation were younger (median age 60.5 years vs 65 years, p < .001) and more likely to have children (63% vs 46%, p < .001). More than half (59%, n = 159) of consultations occurred prior to EP-CT enrollment, while 14% (n = 39) were during patients’ time on EP-CT. The most common reasons for referral on trial included adjustment to illness (41%), limited patient resources (23%), and home/family support (15%). A quarter of referrals (27%, n = 74) were initiated after patients left the trial (27%, n = 74). There were no significant differences in demographic or clinical variables between those referred on EP-CT versus before or after EP-CT. For those who received social work consultation while on EP-CT, median time from date on treatment to consultation was 18 days (0 – 182 days) while median time on EP-CT was 55 days (2 – 576 days). Patients received a median of 2 visits (1 – 20) while on EP-CT. Physicians and research nurses were most likely to refer patients (31% vs 26%, respectively) while 10% of patients self-referred to social work. Conclusions: Over half of EP-CT participants received social work consultations during their cancer course. Most patients who received consultation on an EP-CT did so for psychosocial support. Future research should focus on determining how best to integrate social work into the care of EP-CT participants.

Understanding the supportive care needs of early-phase cancer clinical trial (CT) participants

26

Background: Early phase CTs investigate novel therapeutic approaches for patients with cancer, but little is known about the use of supportive care services among participants in early phase CTs. Methods: We conducted a retrospective chart review of consecutive patients enrolled in Phase 1 CTs from 2017-2019, capturing sociodemographics, clinical data, and use of supportive care services from the electronic health record. We calculated the Royal Marsden Hospital (RMH) prognostic score using data at the time of CT trial enrollment based on patients’ lactate dehydrogenase, albumin, and number of sites of metastasis. The RMH score ranges from 0-3, with scores of 2+ indicating a poor prognosis. We explored differences in patient characteristics, supportive care use, and clinical outcomes based on the RMH prognosis score. Results: Among 426 patients treated on Phase 1 CTs during the study period, the median age was 63.0 years (range 20.5-85.2 years), and most were female (56.1%), white race (85.1%), and had metastatic cancer (97.7%). The most common cancer types were gastrointestinal (22.1%), lung (20.0%), and breast (10.6%) cancer. Under half (31.6%) had an RMH score indicating a poor prognosis. Patients with a poor prognosis score had a worse performance status (ECOG ≥1: 80.2% v 58.3%, p < .001) and more prior treatment (3+ prior lines: 49.5% v 35.0%, p = .001) compared to those without a poor prognosis score. Those with a poor prognosis score were more likely to receive palliative care before or during CT participation (40.5% v 27.1%, p = .011). We observed no significant differences in the rates of nutrition (69.1% v 64.0%), social work (62.2% v 63.8%), or physical therapy (64.5% v 61.7%) consults between those with and without a poor prognosis score. We found that those with an RMH score indicating a poor prognosis had a shorter time on trial (median: 49 vs 87 days, p < .001) and worse survival (median: 139 v 379 days, p < .001). Conclusions: Early phase CT participants represent an advanced cancer population with unique supportive care needs, and we identified a group with a particularly poor prognosis for whom earlier intervention with supportive care services may be needed. Our findings highlight the need to prospectively examine these characteristics along with patient-reported outcomes to better understand the distinct supportive care needs of this population and guide the development of targeted interventions.

FIT: Functional and imaging testing for patients with metastatic cancer

BACKGROUND

Selecting study endpoints in prospective cancer cachexia trials remains poorly defined. The aim of this study was to further evaluate associations in changes in weight, body composition, functional outcomes, and patient-reported outcomes (PROs) in patients with metastatic cancer.

METHODS

We completed a 2-year (2016-2018) observational study in patients with metastatic solid cancer and ECOG performance status 0 to 2 while receiving chemotherapy and/or immunotherapy. We completed assessments at study enrollment and 3 months from enrollment. We analyzed longitudinal changes in weight and body composition using validated methods. Functional assessments included the 6-Min Walk Test, Timed Up and Go Test, and Short Physical Performance Battery. PROs included the Functional Assessment of Anorexia/Cachexia Therapy and Functional Assessment of Cancer Therapy Fatigue. We analyzed changes in body composition and functional assessment using paired t tests. Additionally, we utilized linear regression models to assess relationships between changes in body composition and function outcomes and PROs, adjusting for age and sex.

RESULTS

A total of 57 patients completed baseline assessments, but 19 patients did not complete 3-month assessments (5 died, 1 hospice, 13 withdrew). Of the 38 patients with complete data, the mean age was 61.8 years and 47% were female. Metastatic cancer types included 71% gastrointestinal, 13% lung, and 8% gynecologic. Half received chemotherapy, 16% immunotherapy, and 34% a combination. From enrollment to 3 months, we did not observe a change in weight or skeletal muscle but did find an increase in total adipose tissue (16.9 ± 52.4 cm², 95% CI - 33.79-0.63; p = 0.059; ~ 1.5 pounds). We did not observe any association with changes in weight with any functional outcomes or PROs. However, greater losses in skeletal muscle were associated with greater declines in physical function (6-Min Walk Test [B = 0.04, p = 0.01], Short Physical Performance Battery [B = 2.44, p < 0.01]).

CONCLUSIONS

Patients with metastatic cancer receiving cancer-directed therapy may not experience a change in body weight. However, we found an association between losses in skeletal muscle and greater declines in physical function. Therefore, when selecting study endpoints, prospective cancer cachexia studies may consider selecting changes in body composition over weight.

Relationships among skeletal muscle, symptom burden, health care use, and survival in hospitalized patients with advanced cancer

7006

Background: Loss of skeletal muscle mass (quantity) is common in patients with advanced cancer, but little is known about muscle density (quality). Hospitalized patients with advanced cancer are a highly symptomatic population at risk for the adverse effects of muscle loss. Thus, we sought to describe associations between muscle mass and density, symptom burden, health care use, and survival in these patients. Methods: We prospectively enrolled hospitalized patients with advanced cancer from 9/2014-4/2017. Upon admission, patients reported their physical (Edmonton Symptom Assessment System [ESAS]) and psychological (Patient Health Questionnaire 4 [PHQ4]) symptoms. We used computed tomography (CT) scans performed per routine care ≤45 days prior to enrollment to evaluate muscle mass and density at the level of the third lumbar vertebral body. We categorized patients as sarcopenic using validated sex specific cutoffs. We used regression models to examine associations between muscle mass and density and patients’ symptom burden, health care use, and survival. Results: Of 1,121 patients enrolled, 677 had evaluable CT scan data (mean age = 62.86±12.95 years; 51.1% female). The most common cancer types were gastrointestinal (36.8%) and lung (16.7%) cancer. Most met criteria for sarcopenia (64.0%). Older age and female sex were associated with lower muscle mass (age: B = -0.16, p < .01; female: B = -6.89, p < .01) and density (age: B = -0.33, p < 0.01; female: B = -1.66, p = .01), while higher BMI was associated with higher muscle mass (B = 0.58, p < .01) and lower muscle density (B = -0.61, p < .01). Higher muscle mass was significantly associated with improved survival (HR = 0.97, p < .01), but not with symptom burden or health care use. Higher muscle density was significantly associated with lower ESAS physical (B = -0.17, p = .02), ESAS total (B = -0.29, p < .01), PHQ4 depression (B = -0.03, p < .01) and PHQ4 anxiety (B = -0.03, p < .01) symptoms. Higher muscle density was also associated with decreased hospital length of stay (B = -0.07, p < .01), risk of readmission or death in 90 days (OR = 0.97, p < .01), and improved survival (HR = 0.97, p < .01). Conclusions: Most hospitalized patients with advanced cancer have muscle loss consistent with sarcopenia. We found that muscle mass (quantity) correlated with survival, whereas muscle density (quality) was associated with patients’ symptoms, health care use, and survival. These findings underscore the added importance of assessing muscle quality when seeking to address the adverse effects of muscle loss in oncology.

Patient-reported outcomes (PROs) in older adults with gastrointestinal (GI) cancer undergoing surgery

e24032

Background: Older adults with GI cancer often experience poor surgical outcomes, yet little is known about their PROs, such as physical function, quality of life (QOL), and physical and psychological symptom burden. Methods: As part of a randomized trial of perioperative geriatric care, we prospectively enrolled older adults with GI cancer planning to undergo surgical resection. We asked patients preoperatively to self-report their physical function (ability to perform activities of daily living [ADLs] and instrumental ADLs [IADLs], higher scores indicate better functioning), QOL (EORTC QLQ-C30, higher scores indicate better QOL), symptom burden (Edmonton Symptom Assessment System [ESAS], higher scores indicate more severe symptoms, scores > 3 considered moderate/severe [mod/sev]), and depression symptoms (Geriatric Depression Scale [GDS], higher scores indicate more severe symptoms, scores > 4 represent a positive screen for depression). We used regression models to identify patient characteristics associated with these PROs. We also explored relationships among PROs and surgical outcomes (receiving planned surgery, postoperative readmissions, and survival). Results: We enrolled 160 of 221 (72.4%) patients approached. A minority of patients were independent in all ADLs (5.2%) and IADLs (47.7%). Patients reported an average of 2.56 mod/sev ESAS symptoms, and 27.7% screened positive for depression, with 53.1% reporting at least one comorbidity. The number of comorbidities was significantly associated with impaired ADLs (B = -0.63, P < .01) and lower QOL (EORTC: B = -2.74, P = .03). For surgical outcomes, patients with better physical function were more likely to receive their planned surgery (ADLs: OR = 1.21, P = .02; IADLS: OR = 1.30, P = .03). Higher QOL correlated with greater odds of receiving planned surgery (EORTC: OR = 1.03, P = .06), but this did not reach statistical significance. A higher number of mod/sev ESAS symptoms was associated with greater postoperative readmission risk within 90 days of surgery (HR = 1.13, P = .03). Better physical function was associated with better postoperative survival (ADLs: HR = 0.87, P = .02; IADLs: HR = 0.73, P < .01), and higher depression scores correlated with worse survival (GDS: HR = 1.13, P = .02). Conclusions: Older adults with GI cancer often have baseline functional limitations and a high physical and psychological symptom burden, all of which are associated with worse surgical outcomes. Future work should study whether addressing preoperative PROs could improve older patients’ surgical outcomes. Clinical trial information: NCT02810652 .

Patient-reported care satisfaction and symptom burden in hospitalized patients with cancer

2013

Background: Hospitalized patients with cancer often experience high symptom burden, which may impact their care satisfaction and use of health care services. Yet, studies describing these patients’ care satisfaction, symptom burden, and health care utilization are lacking. Methods: We prospectively enrolled patients with cancer and unplanned hospitalizations from 9/2014-4/2017. Upon admission, patients self-reported their care satisfaction (FAMCARE items asking about satisfaction regarding speed with which symptoms are treated and coordination of care) and physical (Edmonton Symptom Assessment System [ESAS]) and psychological (Patient Health Questionnaire 4 [PHQ4]) symptom burden. We used regression models to identify patient factors associated with care satisfaction. We also explored associations between patients’ care satisfaction, symptom burden, and hospital length of stay (LOS) in models adjusted for age, sex, marital status, comorbidity score, cancer type, cancer documented as curable/incurable, time since cancer diagnosis, and admission to a dedicated oncology service. Results: We enrolled 1,576 of 1,749 (90.1%) consecutive patients (mean age = 63.19±13.39 years, 46.3% female). Most reported being very satisfied/satisfied with the speed with which symptoms are treated (89.0%) and coordination of care (90.1%). Older age (B = 0.01, P < .02 for both) and admission to a dedicated oncology service (B = 0.20, P < .01 for both) were each independently associated with higher satisfaction with the speed with which symptoms are treated and coordination of care. Higher satisfaction with the speed with which symptoms are treated was associated with lower PHQ4 depression (B = -0.14, P = .01), PHQ4 anxiety (B = -0.11, P < .01), ESAS physical (B = -1.30, P < .01), and ESAS total (B = -2.44, P < .01) symptoms. Higher satisfaction with coordination of care was associated with lower PHQ4 depression (B = -0.14, P = .02), PHQ4 anxiety (B = -0.16, P < .01), ESAS physical (B = -1.30, P < .01), and ESAS total (B = -2.75, P < .01) symptoms. Satisfaction with the speed with which symptoms are treated (B = -0.47, P = .03) and coordination of care (B = -0.50, P = .03) were both associated with shorter hospital LOS. Conclusions: Most hospitalized patients with cancer reported high care satisfaction, which was associated with older age and admission to a dedicated oncology service. We found relationships among higher care satisfaction, lower symptom burden, and shorter hospital LOS, underscoring the importance of efforts to enhance symptom management and care coordination in this population.

Randomized trial of a symptom monitoring intervention for hospitalized patients with advanced cancer (NCT03396510)

12014

Background: Hospitalized patients with advanced cancer experience a high symptom burden, which is associated with poor clinical outcomes and increased health care use. Symptom monitoring interventions are increasingly becoming standard of care in oncology, but studies of these interventions in the hospital setting are lacking. We evaluated the impact of a symptom monitoring intervention in hospitalized patients with advanced cancer. Methods: We randomly assigned hospitalized patients with advanced cancer who were admitted to the oncology service to a symptom monitoring intervention or usual care. Patients in both arms reported their symptoms (Edmonton Symptom Assessment System [ESAS] and Patient Health Questionnaire 4 [PHQ4], higher scores on both indicate greater symptom severity) daily via tablet computers. Patients assigned to the intervention had their symptom reports presented graphically with alerts for moderate/severe symptoms during daily oncology rounds. The primary endpoint was the proportion of days with improved symptoms for those who completed two or more days of symptoms. Secondary endpoints included hospital length of stay (LOS) and readmission rates. Results: From 2/2018-10/2019, we randomized 390 patients (76.2% enrollment rate); 320 completed two or more days of symptoms (median age=65.6 [range 18.8-93.2]; 43.8% female). The most common cancers were gastrointestinal (36.9%), lung (18.8%), and genitourinary (12.2%). Nearly half of patients (48.5%) had one or more comorbid conditions in addition to cancer. We found no significant differences between intervention and usual care regarding the proportion of days with improved ESAS total (B=-0.05, P=.17), ESAS physical (B=-0.02, P=.52), PHQ4 anxiety (B=-0.03, P=.33), and PHQ4 depression (B=-0.02, P=.44) symptoms. Intervention patients also did not differ from usual care with respect to secondary endpoints of hospital LOS (7.50 v 7.59 days, P=.88) and readmission rates within 30 days of discharge (32.5% v 25.6%, P=.18). Conclusions: For hospitalized patients with advanced cancer, this symptom monitoring intervention did not have a significant impact on their symptom burden and health care use. These findings do not support the routine integration of this type of symptom monitoring intervention for hospitalized patients with advanced cancer. The positive outcomes seen in previous studies of symptom monitoring interventions may not be reproduced in other patient populations and care settings. Support: UG1CA189823; Clinical trial information: NCT03396510 .

Randomized trial of a perioperative geriatric intervention for older adults with cancer

12012

Background: Older adults with gastrointestinal (GI) cancers undergoing surgery often experience poor outcomes, such as prolonged postoperative (post-op) length of stay (LOS), intensive care unit (ICU) use, and readmissions. Involvement of geriatricians in the care of older adults with cancer can improve outcomes. We conducted a randomized trial of a perioperative geriatric intervention in older adults with GI cancers undergoing surgery. Methods: We randomly assigned patients age ≥65 with GI cancers planning to undergo surgical resection to receive a perioperative geriatric intervention or usual care. Intervention patients met with a geriatrician preoperatively in the outpatient setting and post-op as an inpatient consultant. The geriatrician conducted a geriatric assessment and made recommendations to the surgical/oncology teams. The primary end point was post-op LOS. Secondary end points included post-op ICU use, readmission risk, and patient-reported symptom burden (Edmonton Symptom Assessment System [ESAS]) and depression symptoms (Geriatric Depression Scale). We conducted both intention-to-treat (ITT) and per protocol (PP) analyses. Results: From 9/13/16-4/30/19, we randomized 160 patients (72.4% enrollment rate; median age = 72 [65-92]). The ITT analyses included 137/160 patients who underwent surgery (usual care = 68/78, intervention = 69/82). The PP analyses included the 68 usual care patients and the 30/69 intervention patients who received both pre- and post-op intervention components. In ITT analyses, we found no significant differences between intervention and usual care in post-op LOS (7.2 v 8.2 days, P = .37), ICU use (23.3% v 32.4%, p = .23), and readmission rates within 90 days of surgery (21.7% v 25.0%, p = .65). Intervention patients reported lower depression symptoms (B = -1.39, P < .01) at post-op day 5 and fewer moderate/severe ESAS symptoms at post-op day 60 (B = -1.09, P = .02). In PP analyses, intervention patients had significantly shorter post-op LOS (5.9 v 8.2 days, P = .02) and lower rates of post-op ICU use (13.3% v 32.4%, p < .05), but readmission rates were not significantly different (16.7% v 25.0%, p = .36). Conclusions: Although this perioperative geriatric intervention did not have a significant impact on the primary end point in ITT analysis, we found encouraging results in several secondary outcomes and for the subgroup of patients who received the planned intervention. Future studies of this perioperative geriatric intervention should include efforts, such as telehealth visits, to ensure the intervention is delivered as planned. Clinical trial information: NCT02810652 .

Association of geriatric conditions with survival and health care use in older adults with colon cancer living in long-term care facilities

12045

Background: Older adults with colon cancer residing in nursing homes are at risk for experiencing geriatric conditions such as cognitive decline, limitations in activities of daily living (ADLs), needing pain medications, and incontinence, due to cancer and its treatment. We sought to investigate these factors pre- and post-diagnosis and explored their relationship with health care use and survival. Methods: We identified 483 patients age 65+ with colon cancer from 2011-2015 in SEER-Medicare with linked quarterly nursing home assessments from the Minimum Data Set both pre- and post-cancer diagnosis. We determined the number of geriatric conditions (cognitive functioning, limitation in any ADL, pain medication use, bowel/urinary incontinence) at the pre- and post-cancer diagnosis assessment. We created four groups based on changes in these factors from pre- to post- assessment: improved (n = 105), worsened (n = 25), remained limited (n = 240), never limited (n = 113). Regression models estimated how changes from pre- to post-cancer diagnosis were associated with number of emergency department (ED) visits, hospitalizations, and survival, adjusted for age, sex, race/ethnicity, insurance status, cancer stage, number of pre-cancer comorbidities, urban/rural status, and time from diagnosis. Results: Overall, 55.3% of patients were age > 80 at diagnosis, with 64.8% female; 73.3% non-Hispanic white; and 9.9% Stage IV. Pre- versus post-diagnosis, 20.7% vs. 34.8% of patients were limited in cognitive functioning, and 75.4% vs. 77.8% were limited in ADLs. About a third of patients required pain medication, and about half of patients had urinary incontinence, which did not change pre- and post-diagnosis. Patients who remained limited had higher rates of ED visits (Risk ratio [RR] 1.05, p < .01) compared to those never limited. Those who worsened had higher rate of hospitalization (RR 1.44, p < .01) and ED visits (RR 1.63, p < .01). 12-month and 5-year survival was 46.7% and 6.1%, respectively. Factors associated with worse survival in a multivariable model included: remaining limited at both assessments (OR 1.52, p < .01), worsening from prior (OR 2.00, p = .01), as well as older age and higher cancer stage. Conclusions: Older adults with colon cancer residing in nursing homes have high prevalence of geriatric conditions and differential health care use and survival based on the presence of geriatric conditions, highlighting the need to consider geriatric conditions when providing cancer care to this population.