Search

Objective Hypomagnesemia is associated with diabetes mellitus type 2 and other components of the metabolic syndrome in older patients. Whether serum magnesium concentrations correlate with insulin resistance, obesity, hypertension, dyslipidemia, or other components of metabolic syndrome in younger women with polycystic ovary syndrome (PCOS) is currently unknown. Design Cross-sectional analysis. Setting Academic medical center. Population 100 consecutive women with PCOS by the Rotterdam criteria and 20 age- and BMI-matched normal women. Methods Statistical analysis of the relationship between magnesium levels and a variety of physical, endocrine, and metabolic variables. The STROBE guidelines for a cross-sectional analysis were followed. Main outcome measurements Serum magnesium levels, insulin sensitivity indices, and glucose assessments. BMI, waist circumference, blood pressure, and lipids served as secondary endpoint measurements. Results No patient demonstrated hypomagnesemia. Magnesium levels did not differ between women with and those without insulin resistance, glucose intolerance, or hypertension. Magnesium levels were similar across PCOS phenotypes and WHO-defined BMI categories. Multiple regression analysis did not suggest that serum magnesium concentrations correlated with any physical, metabolic, or endocrine variable. Conclusions Magnesium levels do not correspond with age, BMI, waist circumference, insulin sensitivity, glycemic levels, blood pressure, or lipid levels in reproductive-age women with PCOS. Magnesium concentrations are similar across PCOS phenotypes and indistinguishable from women without PCOS.

BACKGROUND: Patients with malignant gliomas have a poor prognosis and high symptom burden. Caregivers of these patients face significant challenges due to the impact of the cancer on their loved ones neurological and mental faculties. We have previously shown that rates of clinically significant anxiety symptoms in caregivers of this population are high (50%). We sought to evaluate whether caregiving burden was associated with anxiety symptoms in caregivers of patients with malignant gliomas. METHODS: We conducted a prospective study in patients with newly diagnosed malignant gliomas and their caregivers, collecting caregivers self-report data within 6 weeks of diagnosis. We assessed caregivers anxiety symptoms with the Hospital Anxiety and Depression Scale, with subscale scores >7 considered clinically significant. We evaluated caregiving burden using the Caregiver Reaction Assessment (CRA) subscales, with the following domains: impact on schedule, caregivers esteem, lack of family support, impact on health, and impact on finances. We used a linear regression model to identify associations between the CRA subscales and anxiety symptoms. RESULTS: Fifty percent (38/76) of caregivers experienced clinically significant anxiety symptoms. Caregiving burden was significantly associated with anxiety symptoms F(5, 69)=11.525, p

93 Background: Time-driven activity-based costing (TDABC) can be used by health systems to identify inefficiencies and improve the patient experience in clinical encounters. This quality improvement project utilized a Plan, Do, Study, Act (PDSA) cycle to evaluate routine clinic-based care for women with metastatic breast cancer (MBC). Methods: A project plan was developed to directly observe the time spent by MBC patients in clinic (Plan). Patient clinical encounters could include a physician visit along with scans, infusion, and/or labs. We then created process maps of typical patient clinical experiences (Do). Next, we tabulated times (mean, standard deviation [SD]) that patients spent in waiting areas and with each clinical team member (physician, fellow, nurse practitioner, registered nurse, medical assistant, chaplain, social worker, pharmacist, navigator) to identify care inefficiencies (Study). Lastly, we discussed results with providers and identified and implemented strategies for improving efficiency (Act). Results: We directly observed clinic visits (n = 33) for MBC patients from November 2016 to June 2017. On average, patients spent 219 minutes (SD 108) at clinic visits including 71 minutes (SD 45) spent with clinical team members and 85 minutes (SD 43) spent in waiting areas. We identified several opportunities for efficiency improvement, including the delay prior to rooming by medical assistants (n = 31; mean 22, SD 20 minutes), delays with port lab draws in infusion (n = 5; mean 22, SD 13 minutes), and delays awaiting drug from pharmacy (n = 22; mean 15, SD 29 minutes). To improve efficiency, we implemented strategies including having a dedicated infusion nurse assigned to draw labs from patient ports and modifications to medical assistants’ workflow. Conclusions: In this PDSA cycle, we found that patients spend a substantial amount of time at clinic visits, and the majority of this time is spent in waiting areas. Our use of process mapping and evaluation of time spent receiving care identified important opportunities for improving care delivery and efficiency for patients with MBC.

5042Background: Limited data exist regarding disease-related complications (DRCs), such as bone fractures and urinary obstruction (UO), near end of life of men who die with prostate cancer. We aime...

4116Background: FFX is under study in LAPC for its potential for curative resection, but the downstaging rate remains low. Preclinical data suggest that inhibition of the renin-angiotensin system with losartan reduces TGF-b1 activity, enhancing intratumoral penetration of chemotherapy by remodeling desmoplasia and improving perfusion. This study investigated the R0 resection rate of FFX/losartan in LAPC. Methods: LAPC pts (per NCCN criteria), ECOG PS 0-1 were enrolled in a single institution NCI-sponsored phase II study (NCT01821729). Pts received 8 cycles FFX/losartan. If the tumor was radiographically resectable after chemotherapy, pts received short-course chemoradiation (CRT) in 5 fractions (protons 25 GyE, capecitabine 825 mg/m2 bid). If the tumor still abutted vasculature, pts received CRT to 50.4 Gy with a vascular boost to 58.8 Gy. Primary endpoint was R0 resection rate. Secondary endpoints were mPFS, mOS and circulating biomarkers of losartan activity. Results: 50 pts enrolled from 8/2013 to 7/20...

10040Background: As the population ages, it has become increasingly important to understand the risks and benefits of novel agents among older adults with cancer. Immunotherapy (IO) has revolutioni...

6566Background: Cancer patients’ (pts) financial concerns (FCs) represent a barrier to clinical trial (CT) participation and interventions targeting pts’ FCs are needed. We sought to assess the imp...

10012Background: Oncology clinicians often struggle with managing medications and vaccinations in older adults with cancer. We sought to demonstrate the feasibility and preliminary efficacy of inte...

e16089Background: Immune checkpoint inhibition has shown early promising results in patients with chemotherapy-refractory metastatic or advanced tumors of the esophagus, gastroesophageal junction, ...

Importance Childhood cancer survivors may be reluctant to make changes in their employment because of access to health insurance. Objective To examine the prevalence of “job lock” (staying at a job to keep work-related health insurance) in a sample drawn from an established, multi-institutional cohort of full-time employed childhood cancer survivors compared with a random sample of siblings and to explore factors associated with job lock among cancer survivors. Design, Setting, and Participants Cross-sectional survey of full-time employed adult survivors of childhood cancer and a random sample of siblings derived from a cohort of 25 US pediatric oncology centers. Exposures Data collection included sociodemographic factors, insurance coverage, chronic medical conditions, and treatment. Main Outcomes and Measures Self-report of job lock and factors associated with job lock. Results Among the 522 participants, 394 were cancer survivors (54.5% male) and 128 were siblings (51.5% male). Job lock was reported by 23.2% (95% CI, 18.9%-28.1%) of survivors, compared with 16.9% (95% CI, 11.1%-25.0%) of siblings ( P = .16). Job lock was more common among survivors reporting previous health insurance denial (relative risk [RR], 1.60; 95% CI, 1.03-2.52) and problems paying medical bills (RR, 2.43; 95% CI, 1.56-3.80). Among survivors, being female (RR, 1.70; 95% CI, 1.11-2.59; P = .01) and having a severe, disabling, or life-threatening health condition (RR, 1.72; 95% CI, 1.09-2.69; P = .02) were associated with job lock. Conclusions and Relevance Job lock is common among long-term childhood cancer survivors who are employed full-time. A survivor’s decision to remain employed at a job in order to maintain health insurance coverage may affect career trajectory, diminish potential earning power, and ultimately impact quality of life.

This study assessed a randomly selected, age-stratified sample of 698 childhood cancer survivors and 210 siblings and insurance coverage characteristics to determine the impact of underinsurance.

Gait velocity measurements provide functional assessment of patients with diverse diseases and allow predictions about future adverse events. The optimal distance for patient classification is uncertain.Participants were identified in internal medicine clinics and had to be independently ambulatory. Study investigators collected medical information, used a qualitative test to assess gait and balance (G and B score), and measured gait velocity with a timed 100-foot walk.One hundred eighty-four patients participated in this study. The mean age was 57.8 +/- 12.7 years; 50% of the participants were men. The mean gait speed was 3.33 +/- 0.71 ft/sec. Gait speed decreased with age and with body mass index (BMI) and increased with height and male sex. Patients with more comorbidities had decreased speed (P0.01). There were significant correlations between gait speed and grip strength (P0.01) and between lower G and B scores and slower gait speeds (P0.01). G and B scores were negatively correlated with age, BMI, and certain diagnoses. They also predicted risk for past falls. The mean heart rate change during the test was 8 beats per minute. Patients in the highest quartile for heart rate change had lower gait speeds than patients in the other 3 quartiles, suggesting physiologic impairment.A 100-foot walk test in clinic patients provides a practical functional assessment. Gait speed was slower in patients with multiple comorbidities and poor balance. Patients with increased heart rate responses during this test seem to have physiologic impairment. This test has the potential to predict adverse events and to quantitatively determine responses to therapeutic interventions but needs prospective evaluation in clinical studies.

Cancer is a disease occurring disproportionately in older adults. However, the evidence base regarding how best to care for these patients remains limited due to their underrepresentation in cancer clinical trials. Pragmatic clinical trials represent a promising approach for enhancing the evidence base in geriatric oncology by allowing investigators to enroll older, frailer patients onto cancer clinical trials. These trials are more accessible, less resource intensive, and place minimal additional burden on participating patients. Additionally, these trials can be designed to measure endpoints directly relevant to older adults, such as quality of life, functional independence and treatment tolerability which are often not addressed in standard clinical trials. Therefore, pragmatic clinical trials allow researchers to include patients for whom the treatment will ultimately be applied and to utilize meaningful endpoints. Examples of pragmatic studies include both large, simple trials and cluster randomized trials. These study designs allow investigators to conduct clinical trials within the context of everyday practice. Further, researchers can devise these studies to place minimal burden on the patient, the treating clinicians and the participating institutions. In order to be successful, pragmatic trials must efficiently utilize the electronic medical record for data capture while also maximizing patient recruitment, enrollment and retention. Additionally, by strategically utilizing pragmatic clinical trials to test therapies and interventions that have previously shown efficacy in younger, fitter patients, these trials represent a potential mechanism to improve the evidence base in geriatric oncology and enhance care for older adults with cancer.

Clinical trials often represent an increasingly important option for patients with cancer, thus oncologists participating in clinical research need to consider and address the financial burden associated with trial participation. Future research efforts should focus on developing formal screening tools to identify and monitor for financial burden among clinical trial participants.

Predicting when burdensome symptoms will arise or worsen is important to preserving quality of life in patients with serious illness.We explored the relationship between prevalence and severity of symptoms and underlying performance status.We performed a retrospective cohort analysis of patients receiving community palliative care, investigating relationships between symptom burden and performance status. Patient data were obtained from the Carolinas Palliative Care Consortium Database, a central registry of community consultation data for research and quality improvement. We measured symptom prevalence and severity using the McCorkle Symptom Distress Scale and performance status using the Palliative Performance Scale.We analyzed data of 4994 patients, most (90%) with noncancer, serious illnesses. Thirty percent had one or more moderate/severe symptoms. In addition to identifying the high prevalence of fatigue and pain, we found distinct groupings of symptoms with high burden associated with different levels of performance status. This includes high prevalence of fatigue, anorexia, and dyspnea in patients with high performance. Patients with low performance status, however, reported more pain, depression, and constipation.Bothersome symptoms change as patients' performance status worsens. Using performance status as a common language, both medical professionals and informal caregivers can monitor impending changes in symptom burden. This should inform development of community-based delivery systems to detect and manage distress in patients with palliative care needs.

© 2015 Cancer Research UK. All rights reserved. Best supportive care (BSC) as a control arm in clinical trials is poorly defined. We conducted a review to evaluate clinical trials' concordance with published, consensus-based framework for BSC delivery in trials.Methods:A consensus-based Delphi panel previously identified four key domains of BSC delivery in trials: multidisciplinary care; supportive care documentation; symptom assessment; and symptom management. We reviewed trials including BSC control arms from 2002 to 2014 to assess concordance to BSC standards and to selected items from the CONSORT 2010 guidelines.Results:Of 408 articles retrieved, we retained 18 after applying exclusion criteria. Overall, trials conformed to the CONSORT guidelines better than the BSC standards (28% vs 16%). One-third of articles offered a detailed description of BSC, 61% reported regular symptom assessment, and 44% reported using validated symptom assessment measures. One-third reported symptom assessment at identical intervals in both arms. None documented evidence-based symptom management. No studies reported educating patients about symptom management or goals of therapy. No studies reported offering access to palliative care specialists.Conclusions:Reporting of BSC in trials is incomplete, resulting in uncertain internal and external validity. Such studies risk systematically over-estimating the net clinical effect of the comparator arms.

92 Background: In a recent trial, early integrated palliative care (PC) improved quality of life (QOL) and reduced depressive symptoms in patients with newly-diagnosed incurable lung and gastrointestinal (GI) cancer. The mechanisms by which PC benefits these outcomes are unclear. We examined whether early integrated PC improved patients’ coping strategies and the degree to which changes in coping mediated intervention effects on QOL and depressive symptoms. Methods: From 5/2011 to 7/2015, we enrolled 350 patients with newly diagnosed incurable lung or non-colorectal GI cancer in a randomized trial of early PC integrated with oncology care vs. oncology care alone at Massachusetts General Hospital. Patients completed self-report measures of QOL (Functional Assessment of Cancer Therapy-General), depressive symptoms (Patient Health Questionnaire-9), and use of active and avoidant coping strategies (Brief Cope) at baseline, 12, and 24 weeks. Linear regression was used to assess the effects of the early PC intervention on active and avoidant coping strategies. A multiple mediation regression was modeled to examine whether changes in use of coping strategies mediated intervention effects on QOL and depressive symptoms. Results: Compared to oncology care, early integrated PC improved patients’ use of active coping ( B= 1.09, SE= 0.44, p= .01, 95% CI: 0.23, 1.96) and slightly reduced use of avoidant coping strategies ( B= -0.44, SE= 0.23, p= .06, 95% CI:-0.90, 0.02). Improvements in 24-week QOL in patients assigned to PC were mediated by increased use of active coping strategies (indirect effect = 1.27, 95% CI: 0.31, 2.86), but not by decreased use of avoidant coping. Similarly, PC-related reductions in 24-week depressive symptoms were mediated by increased use of active coping strategies (indirect effect = -0.39, 95% CI:-0.89, -0.08) but not by decreased use of avoidant coping. Conclusions: Patients with newly diagnosed incurable cancer who received early integrated PC showed increased use of active coping strategies, which was associated with improved QOL and depressive symptoms. PC may improve QOL and mood by providing patients with the skills to cope effectively with life-threatening illness. Clinical trial information: NCT01401907.

10050 Background: Patients with cancer often experience depression, which is associated with worse outcomes, including longer hospital length of stay (LOS). Although antidepressant medication can improve depressive symptoms in patients with cancer, it is unclear whether their use translates into better outcomes. We sought to clarify the relationship between depressive symptoms, antidepressant medication, and hospital LOS in patients with advanced cancer. Methods: We enrolled hospitalized patients with advanced cancer from 9/2014 to 4/2016 as part of a longitudinal data repository. We examined patients’ medical records to obtain information about documented depressive symptoms in the 3 months prior to admission and use of antidepressant medication at the time of admission. Using descriptive statistics, we compared differences in patient characteristics and hospital LOS across these groups. We used linear regression to examine associations and moderation effects between depressive symptoms, use of antidepressant medication, and hospital LOS. Results: Of 1,036 enrolled patients (89.9% of approached), 126 (12.2%) had documented depressive symptoms in the 3 months prior to admission and 288 (27.8%) were taking an antidepressant medication at the time of admission. Patients with depressive symptoms were more likely to be on antidepressant medication at admission than those without depressive symptoms (48.4% vs 24.9%, p < .001). Patients taking antidepressant medication were younger (62.4 vs 64.4 years, p = .026) and more likely to be female (55.2% vs 47.2%, p = .021). Depressive symptoms were associated with longer hospital LOS (7.3 vs 6.1 days, p = .036), and antidepressant medication was a moderator of this relationship. Among patients not on antidepressant medication, depressive symptoms were associated with longer hospital LOS (7.9 vs 6.1 days, p = .025), but among those on antidepressant medication, depressive symptoms were not associated with hospital LOS (6.6 vs 6.2 days, p = .588). Conclusions: Antidepressant medication moderated the relationship between depressive symptoms and longer hospital LOS. Our results support the need to recognize and address depressive symptoms in patients with advanced cancer.

6504 Background: Patients with advanced cancer experience frequent hospitalizations and burdensome transitions of care post-discharge. We examined predictors of discharge location for patients with advanced cancer. Methods: We prospectively enrolled patients with advanced cancer with unplanned hospitalizations from 9/14 to 3/16. Upon admission, we used the Edmonton Symptom Assessment Scale and Patient Health Questionnaire-4 to assess physical and psychological symptoms, respectively. We used logistic regression models to identify predictors of discharge to location other than home, including post-acute care (PAC) [skilled nursing facility or long term acute care hospital] or hospice [any setting]. We used Cox regression models adjusted for clinical variables to assess the relationship between discharge location and survival. Results: Out of 932 patients, 726 (77.9%) were discharged home, 118 (12.7%) to PAC and 88 (9.4%) to hospice. Compared with patients discharged home, those discharged to PAC or hospice had higher symptom burden, including dyspnea, constipation, low appetite, drowsiness, fatigue, depression, and anxiety (all p < 0.05). Patients discharged to PAC or hospice vs. home were more likely to be older (OR 1.03, p < 0.0001), live alone (OR 1.95, 95% CI: 1.25-3.02, p < 0.003), have impaired mobility (OR 5.08, 95% CI: 3.46-7.45, p < 0.0001), longer length of stay (OR 1.15, 95% CI: 1.11-1.20, p < 0.0001), higher ESAS physical symptoms (OR 1.02, 95% CI: 1.003-1.032, p < 0.017), and higher PHQ-2 depression symptoms (OR 1.13, 95% CI: 1.01-1.25, p < 0.027). Patients discharged to hospice vs. PAC were more likely to receive palliative care consultation (OR 4.44, 95% CI: 2.12 to 9.29, p < 0.0001) and have shorter length of stay (OR 0.84, 95% CI: 0.77 to 0.91, p < 0.0001). Compared with patients discharged home, those discharged to PAC had lower survival (HR 1.53, 95% CI 1.22-1.93, p < 0.0001). Conclusions: Patients with advanced cancer discharged to PAC or hospice have substantial physical and psychological symptom burden and poor physical function. Patients discharged to PAC also have inferior survival compared with those discharged home. They may benefit from targeted interventions to improve their quality of life and care.

6579 Background: Prolonged hospital admissions are often inconsistent with patients’ preferences and incur significant costs. While patients’ symptoms may result in hospitalizations, the relationship between patients’ symptom burden and their hospital length-of-stay (LOS) has not been fully explored in patients with curable cancers. Methods: We prospectively enrolled patients with curable cancer and unplanned hospital admissions between 8/2015 and 12/2016. Within the first 5 days of admission, we assessed patients’ physical (Edmonton Symptom Assessment System [ESAS]; scored 0-10 with higher scores indicating greater symptom burden) and psychological symptoms (Patient Health Questionnaire 4 [PHQ-4]; scored categorically and continuous with higher scores indicating greater distress). We created summated ESAS total and physical symptom variables. To assess the relationship between patients’ symptom burden and their hospital LOS, we used separate linear regression models adjusted for age, sex, marital status, education level, time since cancer diagnosis, and cancer type. Results: We enrolled 452 of 497 (91%) approached patients (mean age = 61.9 years; 188 [42%] female). Over half had hematologic cancers (n = 249, 55%). Mean hospital LOS was 8.3 days. Over one-tenth of patients screened positive for PHQ-4 depression (n = 74, 16%) and anxiety (n = 60, 13%) symptoms. Mean ESAS symptom scores were highest for fatigue (6.6), drowsiness (5.4), pain (4.9), and lack of appetite (4.8). In multivariable regression analysis, patients’ physical and psychological symptoms were associated with longer hospital LOS (table). Conclusions: Patients with curable cancer and unplanned hospital admissions experience a substantial symptom burden, which predicts for prolonged hospitalizations. Importantly, patients’ symptoms are modifiable risk factors that, if properly addressed, can improve care delivery and may have the potential to help decrease prolonged hospitalizations. [Table: see text]

4117 Background: With the advent of more effective therapies for metastatic pancreatic ductal adenocarcinoma (PDAC), efforts to incorporate these agents, such as FOLFIRINOX, into the neoadjuvant setting are increasing. However, the efficacy and cost-effectiveness of using neoadjuvant FOLFIRINOX for patients with borderline resectable or locally advanced PDAC are unknown. We performed a decision analysis to assess the value of neoadjuvant FOLFIRINOX versus upfront surgery and adjuvant therapy. Methods: We developed a mathematical simulation model to evaluate the efficacy and cost-effectiveness of neoadjuvant FOLFIRINOX compared to upfront surgery and adjuvant therapy. We used published and institutional data as inputs to inform model development. Model outcomes included overall and disease-free survival, net benefits expressed as discounted quality-adjusted life-years (QALYs), costs in US dollars, and cost-effectiveness expressed as an incremental cost-effectiveness ratio. We used deterministic and probabilistic sensitivity analyses to explore the uncertainty of model assumptions. Results: Model estimated median overall survival (29 vs 23 months) and disease-free survival (14 vs 13 months) were better for neoadjuvant strategy compared with upfront surgery. Neoadjuvant strategy resulted in an additional 0.68 life-years gained, or 0.57 QALYs, at a cost of $59,000/QALY gained. Sensitivity analysis found that cancer recurrence rates affected model results the most. Our findings were otherwise robust with respect to changes in other model parameters, including chemotherapy toxicity, surgical complications and cancer mortality. Probabilistic sensitivity analyses showed that neoadjuvant strategy was cost-effective 80% of the time with a willingness-to-pay threshold of $100,000/QALY. Conclusions: Our model results demonstrate that neoadjuvant strategy is preferable to upfront surgery for patients with borderline resectable or locally advanced PDAC from both an efficacy and cost-effectiveness standpoint. Additional clinical data are needed to further define the long-term effectiveness of neoadjuvant FOLFIRINOX to confirm our results.

10040 Background: Cancer disproportionately affects older adults, yet research defining the supportive care needs of these patients is lacking. We sought to examine associations between geriatric impairments, quality of life (QOL), and physical and psychological symptom burden in older adults with newly diagnosed incurable gastrointestinal (GI) cancer. Methods: We prospectively enrolled patients age ≥70 within 8 weeks of diagnosis of incurable GI cancer at Massachusetts General Hospital from 10/2015-11/2016. We used surveys to assess geriatric impairments (Vulnerable Elders Survey-13 [range 0-10, scores ≥3 identify patients with impairments), QOL (EORTC QLQ-C30 [range 0-100, higher scores indicate better QOL]), physical symptoms (Edmonton Symptom Assessment System [range 0-10, higher scores indicate greater symptom burden]) and psychological symptoms (Geriatric Depression Scale, [range 0-15, higher scores indicate greater depression symptoms]). We used descriptive statistics to determine differences in patient characteristics by the presence or absence of geriatric impairments. We used linear regression adjusted for age, employment, cancer type, and comorbidity to examine associations between geriatric impairments, QOL, and physical and psychological symptom burden. Results: We enrolled 50 of 58 (86%) patients approached (mean age = 78.7; 52% with pancreatic cancer). Nearly half (46%) screened positive for geriatric impairments; these patients were older (81.7 vs 76.1, p < .01) and had more comorbid conditions (2.4 vs 1.2, p = .01). On linear regression, patients with geriatric impairments reported worse QOL across all domains (General QOL: B = -28.3, p < .01; Physical: B = -36.8, p < .01; Role: B = -36.8, p < .01; Emotional: B = -30.1, p < .01; Cognitive: B = -17.8, p = .03; Social: B = -39.7, p < .01), higher depression scores (B = 5.1, p < .01) and worse fatigue (B = 4.6, p < .01), drowsiness (B = 4.0, p < .01), appetite (B = 3.8, p < .01), and pain (B = 2.7, p = .02). Conclusions: Older adults with advanced cancer experience considerable unmet supportive care needs, particularly those with geriatric impairments. Future research is needed to assess older patients for geriatric impairments and address their unique palliative and supportive care needs.

6592 Background: Code status discussions ensure the delivery of preference-concordant care. However, the processes by which hospitalized patients with advanced cancer change their code status from full code to DNR are unknown. Methods: We conducted a mixed-methods study on a prospective cohort of patients with advanced cancer who were hospitalized from 9/14-10/15. Two physicians used a consensus-driven medical record review to characterize processes leading to code status transitions from full code to DNR. We explored factors associated with these processes using χ2 and Kruskal-Wallis tests. Results: We reviewed 1,047 hospitalizations among 728 patients. Admitting physicians did not address code status in 52.1% of these hospitalizations, leading code status orders to be “presumed full.” 273 patients (37.5%) transitioned from full code to DNR; 132 (48.4%) of them had erroneous “presumed full” code status orders on admission. We identified three additional processes leading to transitions from full code to DNR: acute clinical deterioration (15.4%), discontinuation of cancer-directed therapy (17.2%), and hypothetical discussions regarding the futility of CPR (15.4%). Among these processes, code status transitions due to acute clinical deterioration were associated with less patient involvement, shorter time to death, and higher likelihood of inpatient death. Changes due to hypothetical discussions were more likely to involve palliative care. Conclusions: Half of code status transitions among hospitalized patients with advanced cancer were due to erroneous full code orders, underscoring a greater need to discuss patient CPR preferences. Transitions due to acute clinical deterioration were associated with less patient engagement and higher likelihood of inpatient death. [Table: see text]

10023 Background: In a recent trial, early integrated palliative care (PC) improved quality of life (QOL) and reduced depression symptoms in patients with newly-diagnosed incurable lung and gastrointestinal (GI) cancer. The mechanisms by which PC benefits these outcomes are unclear. Therefore, we examined whether early integrated PC improved patients’ coping strategies and the degree to which changes in coping mediated intervention effects on patient-reported QOL and depression symptoms. Methods: From 5/2011 to 7/2015, we enrolled 350 patients with newly diagnosed incurable lung or non-colorectal GI cancer in a randomized trial of early PC integrated with oncology care vs. oncology care alone at Massachusetts General Hospital. Patients completed self-report measures of QOL (Functional Assessment of Cancer Therapy-General), depression symptoms (Patient Health Questionnaire-9), and use of active and avoidant coping strategies (Brief Cope) at baseline, 12, and 24 weeks. Linear regression was used to assess the effects of the early PC intervention on active and avoidant coping strategies. A multiple mediation regression with bias-corrected bootstrapping was modeled to examine whether changes in use of coping strategies mediated intervention effects on QOL and depression symptoms. Results: Compared to oncology care, early integrated PC improved patients’ use of active coping and reduced use of avoidant coping strategies. Improvements in 24-week QOL in patients assigned to PC were mediated by increased use of active coping strategies (indirect effect = 1.27, bootstrapped SE = 0.65, 95% CI [0.31, 2.86]), but not by decreased use of avoidant coping. Similarly, PC-related reductions in 24-week depression symptoms were mediated by increased use of active coping strategies (indirect effect = -0.39, bootstrapped SE = 0.20, 95% CI [-0.89, -0.08] but not by decreased use of avoidant coping. Conclusions: Patients with newly diagnosed incurable cancer who received early integrated PC showed increased use of active coping strategies, which led to improved QOL and depression symptoms. PC may improve QOL and mood by providing patients with the skills to cope effectively with life-threatening illness. Clinical trial information: NCT01401907.

e18078 Background: The digital divide has shifted from disparities in internet access to disparities in content and design. Technology overlooks limited English proficient (LEP) patients, resulting in a lack of translated online content, possibly increasing disparities in cancer prevention, treatment and clinical trial participation. We sought to determine the language accessibility for the websites of the NCI-Designated CC. Methods: In January 2017, we performed a cross-sectional review of the language accessibility of NCI-Designated CC homepages using manual review and informatics methods (i.e., web scraping). Web scraping automates data extraction from online content. The primary outcomes were presence of translated content, number of languages available and method of translation defined as no translation, Google Translate (GT) or manual translation. Manual translation was categorized as limited (few phrases), moderate (1 webpage) or full ( > 1 webpage or entire site). We used logistic regression to assess the relationship between translated website content and CC county demographics: percent LEP, median income and percent of households with an internet subscription. We performed Spearman Rank Correlation by ranking translation effort: no translation, GT, limited, moderate and full translation. Results: Of 69 NCI-Designated CC websites, 54 (78.3%) were without translation, 12 (17.4%) were manually translated and 3 (4.3%) used GT. Of 12 manually translated websites, 6 had fully, 4 had moderate and 2 had limited translations. Of 16 languages offered, Spanish was the most common (100%), followed by Chinese (50%) and Arabic (33%). We found no significant increase in the odds of having a translated website as related to LEP population, median income or percent of households with internet subscription by CC county. There was no correlation between the translation effort and LEP population, internet subscription or median income by CC county. Conclusions: We found that most NCI-Designated CC offered no translations of their website content. Despite cancer health disparities and the increasing role of health technology, the NCI-Designated CC websites currently remain inaccessible to LEP patients.

e18071 Background: Age, sex, and racial/ethnic disparities exist for several malignancies, but these relationships are understudied in pancreatic adenocarcinoma (PDAC). We used the Surveillance, Epidemiology, and End Results (SEER)-Medicare linked database to determine whether these disparities persist in survival and treatment after adjusting for demographic, clinical, and treatment characteristics. Methods: Our study includes PDAC patients (pts) diagnosed from 1992-2011 in the SEER-Medicare database. We used Cox regression models to compare survival across age, sex, and race/ethnicity, stratified by non-metastatic and metastatic cancer. We used logistic regression to compare differences in chemotherapy, radiation, and surgery receipt across age, sex, and race/ethnicity. We adjusted for marital status, urban location, socioeconomic factors, SEER region, comorbidities, cancer stage, lymph node status, tumor location, tumor grade, diagnosis year, and chemotherapy, radiation, and/or surgery receipt. Results: Of 20,896 PDAC pts (median diagnosis age = 75 years; 56% female), 84% were White, 9% Black, 5% Asian, and 2% Hispanic; 53% had metastatic cancer. Among non-metastatic pts, adjusted Cox regression demonstrated that older pts had worse survival compared with younger pts (HR: ≥1.1 for all ages [reference = 66-69 years vs. 70-74, 75-79, 80-84, 85+], p < 0.01 for all); we found no survival differences between sexes. Black (HR: 1.1, p = 0.01) and Hispanic (HR: 1.2, p < 0.01) pts had worse survival compared to White pts. Among metastatic cancer pts, adjusted Cox regression demonstrated that older pts (HR: 1.1 for age 85+ [reference = 66-69 years], p < 0.01) had worse survival than younger pts, and males (HR: 1.1, p < 0.01) had worse survival than females. There were no racial/ethnic differences. There were disparities in treatment received; older age and minority race/ethnicity were associated with lower likelihood of receiving chemotherapy, radiation, and/or surgery. Conclusions: Age, sex, and racial/ethnic disparities in survival outcomes and treatment received exist for PDAC pts; these disparities persist after adjusting for differences in demographic, clinical, and treatment characteristics.

Background Cancer patients may experience financial distress as a side effect of their care. Little is known about which patients are at greatest risk for altering their care or lifestyle due to treatment-related financial distress. Methods We conducted a cross-sectional survey study to determine which patients are at greatest risk for altering their care or lifestyle due to treatment-related financial distress. Eligible patients were adults receiving cancer treatment enrolled between June 2010 and May 2011. We grouped coping strategies as lifestyle altering or care altering. We assessed coping strategies and relationships between covariates using descriptive statistics and analysis of variance. Results Among 174 participants, 89% used at least one lifestyle-altering coping strategy, while 39% used a care-altering strategy. Care-altering coping strategies adopted by patients included the following: not filling a prescription (28%) and taking less medication than prescribed (23%). Lifestyle-altering strategies included the following: spending less on leisure activities (77%), spending less on basics like food and clothing (57%), borrowing money (54%), and spending savings (50%). Younger patients were more likely than older patients to use coping strategies (p

Background. Recent literature has associated hepatitis C virus with the development of non-Hodgkin lymphoma. Hepatitis C virus infection appears to promote lymphoproliferation, providing a plausible mechanism for a causative association; however, despite prior reports of patients with comorbid hepatitis C infection and Waldenström macroglobulinemia, the literature is in disagreement regarding whether there exists an association between these two conditions.Case Presentation. This case report describes a 57-year-old African-American male with chronic hepatitis C infection and cryoglobulinemia who presented with several episodes of transient confusion and paralysis and was found to have symptomatic hyperviscosity. The recognition of his condition was facilitated by characteristic findings on ophthalmologic examination. He was subsequently diagnosed with Waldenström macroglobulinemia on bone marrow biopsy.Conclusions. An up to date, comprehensive review of the literature suggests an association between hepatitis C and Waldenström macroglobulinemia. Data on optimal treatment of patients with comorbid hepatitis C infection and Waldenström macroglobulinemia is limited. We have provided a comprehensive review of previously explored treatment options to guide management of other similar patients. Our patient has since been treated with repeated plasmapheresis with a plan to pursue antiviral therapy.

The cancer anorexia-cachexia syndrome (CACS) is common in patients with advanced solid tumors and is associated with adverse outcomes including poor quality of life (QOL), impaired functioning, and shortened survival.To apply the recently posed weight-based international consensus CACS definition to a population of patients with advanced non-small cell lung cancer (NSCLC) and explore its impact on patient-reported outcomes.Ninety-nine patients participated in up to four study visits over a six-month period. Longitudinal assessments included measures of physical function, QOL, and other clinical variables such as weight and survival.Patients meeting the consensus CACS criteria at Visit 1 had a significantly shorter median survival (239.5 vs. 446 days; hazard ratio, 2.06, P 0.05). Physical function was worse in the CACS group (mean Karnofsky Performance Status score 68 vs. 77, Eastern Cooperative Oncology Group Performance Status score 1.8 vs. 1.3, P 0.05 for both), as was QOL (Functional Assessment of Cancer Therapy-General [FACT-G] Lung Cancer subscale of 17.2 vs. 19.9, Anorexia/Cachexia subscale of 31.4 vs. 37.9, P 0.05 for both). Differences in the FACT-G and the Functional Assessment of Chronic Illness Therapy-Fatigue subscale approached but did not reach statistical significance. Longitudinally, all measures of physical function and QOL worsened regardless of CACS status, but the rate of decline was more rapid in the CACS group.The weight-based component of the recently proposed international consensus CACS definition is useful in identifying patients with advanced NSCLC who are likely to have significantly inferior survival and who will develop more precipitous declines in physical function and QOL. This definition may be useful for clinical screening purposes and identify patients with high palliative care needs.

The decision to treat with chemotherapy in advanced cancers is a complex process that requires oncologists to weigh the indications and benefits. Treatment of patients with advanced cancer is typically multidimensional and should ideally include oncologists in coordination with palliative care specialists. When the intent of chemotherapy for patients with advanced neoplasms is no longer curative, palliative care can and should be delivered simultaneously with antineoplastic agents. Often, chemotherapy for advanced cancers is delivered in an attempt to palliate symptoms, and therefore improve quality of life (QOL). The role of palliative chemotherapy should be continually reevaluated throughout the trajectory of the patient's illness. When chemotherapy is no longer controlling the disease or helping symptoms, the role of chemotherapy should be reconsidered. This review aims to provide a foundation for discussions about treatment of patients with advanced malignancies.

Acute myeloid leukemia (AML) is an aggressive disease that predominantly affects elderly patients. Cytokines and chemokines are major players in the pathogenesis of AML. They regulate the disease course and play a deleterious role in the progression of AML. The geriatric population is particularly vulnerable to these mediators as these cytokines and chemokines are also implicated in the development of frailty, fatigue, and declining cognitive function. It is the combination of these adverse effects of cytokines and chemokines that affect performance status and, in turn, the poor prognosis in this age group. Cytokines and chemokines are emerging as therapeutic targets in AML. Future endeavors to treat AML will likely involve cytokines and chemokines as attempts are made to disrupt the bone marrow environment. By modulating the bone marrow stroma, the goal is to create an environment less favorable to AML cells and more favorable to the effects of chemotherapy against AML.

234 Background: Patients with newly diagnosed advanced cancers who receive early palliative care (PC) integrated with oncology care report improvement in their quality of life (QOL) and mood. While a telephone-based psycho-educational intervention for FC decrease depression, the impact of early, integrated outpatient PC models on FC outcomes remains unknown. Methods: We conducted a randomized trial of early PC integrated with oncology care versus oncology care alone for newly diagnosed patients with incurable lung and gastrointestinal cancers and their FC. Eligible FC were relatives or friends who would likely accompany the patient to clinic visits. FC were eligible to enroll up to 4 weeks after patient enrollment. The intervention entailed at least monthly patient visits with PC from the time of enrollment. FC were not required to attend these visits. We used the Medical Outcomes Study Health Survey Short Form-36 to examine QOL and the Hospital Anxiety and Depression Scale to examine mood at baseline, weeks 12 and 24. We used the two-sample t-test to examine changes in QOL and mood from baseline to week 12 and week 24. Results: We enrolled 350 patients and 275 (78.6%) of potentially eligible FC (control n = 138, intervention n = 137) between 5/2/2011 to 7/20/2015. FC outcomes missing data rate at 12 and 24 weeks were 16.8% and 33.8% respectively. At 12 weeks, FC of patients receiving early PC reported higher vitality (1.1 vs. -3.2, p = 0.05) and social functioning (-3.0 vs. -3.8, p = 0.02), and lower depression symptoms (-0.45 vs. 0.24, p = 0.03) compared to FC of patients assigned to usual care. At 24 weeks, FC of patients assigned to early PC had lower depression symptoms (-0.37 vs. 0.28, p = 0.05), but no differences in vitality or social functioning compared to those receiving usual care. No differences in other SF-36 subscales or anxiety were noted at 12 and 24 weeks. Conclusions: Early involvement of PC for patients with newly diagnosed lung and gastrointestinal cancers leads to improvements in FC depression and aspects of QOL. This work demonstrates that the benefits of early, integrated palliative care models in oncology care extend beyond patient outcomes and positively impact the experience of FC. Clinical trial information: NCT01401907.

98 Background: Patients with incurable cancer are often hospitalized and have frequent readmissions after discharge. Considering the high physical and psychological symptom burden in this population, we sought to investigate symptoms as predictors of hospital length of stay (LOS) and time to first unplanned readmission. Methods: We consecutively enrolled incurable cancer patients with unplanned hospital admissions from 9/2014-4/2016. Within the first 5 days of admission, we assessed physical (Edmonton Symptom Assessment System [ESAS]; scored 0-10) and mood symptoms (Patient Health Questionnaire 4 [PHQ-4]; scored categorically). We created summated ESAS total and physical symptom variables. To identify predictors of LOS we used linear regression and for time to readmission we used Cox regression, with all models adjusted for age, sex, marital status, comorbidity, education, cancer type and time since incurable diagnosis. Results: We enrolled 1,000 of 1,227 (81%) eligible patients (mean age = 63.4; 50% female; 66% married). Gastrointestinal (33%) and lung (18%) cancers were the most common. Mean hospital LOS was 6.2 days and 30-day readmission rate was 25%. Over half of patients reported moderate/severe fatigue, drowsiness, lack of appetite, pain and poor well-being. Over one-fourth screened positive for PHQ depression and anxiety. All physical and mood symptoms individually predicted for longer LOS. Pain, nausea, poor well-being, ESAS total, ESAS physical and PHQ anxiety predicted for shorter time to readmission. Conclusions: Hospitalized patients with incurable cancer experience a high symptom burden, which correlates with their health care utilization. Both physical and psychological symptoms predict for longer hospital LOS and shorter time to readmission. These findings can inform interventions targeting patients’ symptoms during hospital admissions in an effort to improve health care delivery and utilization. [Table: see text]

10003Background: Early palliative care (PC) improves outcomes in patients with newly diagnosed metastatic non-small cell lung cancer (NSCLC) and in patients identified by clinicians as having poor prognosis, advanced cancer. We evaluated the impact of early, integrated palliative and oncology care in patients with newly diagnosed lung and gastrointestinal (GI) cancer. Methods: We randomly assigned patients with newly diagnosed incurable lung (NSCLC, small cell, mesothelioma) or GI (pancreas, hepatobiliary, gastric, esophageal) cancer to PC integrated with oncology care (at least monthly visits with PC) or usual oncology care. We used the Functional Assessment of Cancer Therapy-General (FACT-G) to assess quality of life (QOL) and the Patient Health Questionniare-9 (PHQ-9) and Hospital Anxiety and Depression Scale (HADS) for mood at baseline, weeks 12 and 24. We assessed patients’ perception of their likelihood of cure and their communication about their end of life (EOL) preferences. WΒe used linear regres...

10123Background: Patients with incurable cancer often require hospitalizations for symptom management, but the relationship between patients’ symptom burden and their hospital LOS is unknown. We so...

e18056Background: Clinical trials (CTs) often represent the best option for patients (pts) with cancer, yet the financial burden (FB) of CT pts is understudied. We sought to assess the FB of CT pts...

10131Background: Patients with newly diagnosed advanced cancers who receive early palliative care (PC) integrated with oncology care report improvement in their quality of life (QOL) and mood. Whil...

6535Background: Cancer survivors represent a population that requires affordable and accessible long-term medical care, yet studies examining their financial burden and health care experience are l...

20 Background: Survivors of childhood cancer face increased risks for morbidity and premature mortality due to the sequelae of their primary disease and its treatment. Survivors may be socioeconomically vulnerable, which could negatively impact their healthcare utilization. We investigated sociodemographic factors associated with forgoing needed healthcare due to cost among long-term survivors within the CCSS. Methods: From a survey of insured and uninsured survivors conducted in 2011-12, forgoing needed healthcare due to cost was determined by the question “In the past year, was there a time when you did any of the following because you were worried about the cost? (Yes/No)” with 10 response categories (e.g., skipping a test/treatment, postponing medical care, taking a smaller cost of a prescription). ‘Yes’ responses were summarized using a categorical outcome variable of 0, 1-2, or ≥ 3 instances of forgoing healthcare. Ordinal logistic regression models assessed the association of sociodemographic factors with increased instances of forgoing healthcare due to cost, including survey weights to account for stratified sampling. Results: Of 1,110 mailed surveys, we received 698 (64%) responses. Mean age at time of survey was 39.6 (range 24-60) years and average time since diagnosis was 31.3 (SD = 4.6) years. 23% of survivors reported 1-2 instances of foregoing needed healthcare and 31% reported 3 or more. Odds of more forgone needed healthcare due to costs were increased for those who were uninsured, female, and had a chronic medical condition. [Table: see text]

Hypomagnesemia is associated with diabetes mellitus type 2 and other components of the metabolic syndrome in older patients. Whether serum magnesium concentrations correlate with insulin resistance, obesity, hypertension, dyslipidemia, or other components of metabolic syndrome in younger women with polycystic ovary syndrome (PCOS) is currently unknown.Cross-sectional analysis.Academic medical center.100 consecutive women with PCOS by the Rotterdam criteria and 20 age- and BMI-matched normal women.Statistical analysis of the relationship between magnesium levels and a variety of physical, endocrine, and metabolic variables. The STROBE guidelines for a cross-sectional analysis were followed.Serum magnesium levels, insulin sensitivity indices, and glucose assessments. BMI, waist circumference, blood pressure, and lipids served as secondary endpoint measurements.No patient demonstrated hypomagnesemia. Magnesium levels did not differ between women with and those without insulin resistance, glucose intolerance, or hypertension. Magnesium levels were similar across PCOS phenotypes and WHO-defined BMI categories. Multiple regression analysis did not suggest that serum magnesium concentrations correlated with any physical, metabolic, or endocrine variable.Magnesium levels do not correspond with age, BMI, waist circumference, insulin sensitivity, glycemic levels, blood pressure, or lipid levels in reproductive-age women with PCOS. Magnesium concentrations are similar across PCOS phenotypes and indistinguishable from women without PCOS.

139 Background: Cancer clinical trials often enroll patients with advanced cancer, yet trial participation should not preclude the appropriate use of palliative care (PC) services for these patients. We sought to describe rates and correlates of PC consultation in clinical trial patients and identify factors associated with use of PC. Methods: We prospectively collected data on patient characteristics and rates of PC consultation among all patients enrolled in cancer clinical trials at Massachusetts General Hospital in 2014. We used multiple logistic regression with purposeful selection of covariates to identify factors associated with PC consultation. Results: Of 1,218 clinical trial patients (mean age = 58 years; 643 (53%) female), 781 (64%) were married and 851 (70%) had metastatic disease. All cancer types were represented, but hematologic cancers (21%) were most common. Of those who had tumor genotype testing, 340 (62%) had a somatic mutation. Overall, 177 (15%) received PC consultation. Of those with metastatic disease, 149 (18%) received a PC consult. Metastatic disease, being unmarried, and absence of a somatic tumor mutation correlated with receipt of a PC consult. Conclusions: We found that a minority of cancer clinical trial patients receive PC consultation. Having metastatic disease and being unmarried correlated with receipt of PC consultation, thus highlighting a population likely needing additional supportive care. Presence of a somatic mutation was inversely associated with receiving PC consultation, suggesting that disease biology may impact patients’ supportive care needs or presence of these mutations influences clinician decisions about the need for PC services. These data help identify the factors associated with appropriate use of PC, and add to our understanding about the use of these services in cancer clinical trial patients. [Table: see text]

100 Background: Patients with advanced cancer experience high rates of both physical and psychological morbidity, but data describing patients’ symptoms during hospital admissions are lacking. We sought to describe symptom burden in hospitalized patients with incurable solid and hematologic malignancies. Methods: We prospectively enrolled patients with incurable cancers admitted to the Massachusetts General Hospital from 9/1/2014 through 5/1/2015. Within the first week of their admission, we assessed physical and psychological symptoms using the Edmonton Symptom Assessment System-revised (ESAS-r). Beginning 11/15/2015, we also administered the Patient Health Questionnaire 4 (PHQ-4), scored categorically. Results: We enrolled 457 of 547 (84%) eligible patients. Participants (mean age=63.8 years; n=231, 51% female) had the following malignancies: gastrointestinal (n=149, 33%), lung (n=77, 17%), genitourinary (n=52, 11%), breast (n=33, 7%), hematologic (n=24, 5%), and other solid tumors (n=122, 27%). Using the ESAS-r, tiredness, drowsiness, anorexia, and pain were the most common severe symptoms. Using the PHQ-4, approximately one-third of participants screened positive for depression (91/271, 34%) and anxiety (86/273, 32%). Conclusions: Hospitalized patients with incurable solid and hematologic malignancies experience substantial physical and psychological symptoms. Most patients reported at least moderate tiredness, drowsiness, anorexia and pain. Additionally, a concerning proportion reported depression and anxiety. Our data demonstrate the need for efforts to alleviate the physical symptoms experienced by this population, while also seeking to understand and address their psychological needs. [Table: see text]

160 Background: Cancer clinical trials (CTs) often represent the best available treatment for many patients, but little is known about the health care utilization of these patients. We examined correlates of hospitalizations and emergency department (ED) use in cancer CT patients to determine those at greatest risk for these outcomes. Methods: We prospectively collected data on patient characteristics, hospitalizations and ED use among all patients enrolled in cancer CTs at Massachusetts General Hospital in 2014. We calculated the number of hospitalizations and ED visits in the 6-month interval following patients’ CT enrollment. We used linear regression with purposeful selection of covariates to identify factors associated with hospitalizations and ED use. Results: Of 1,218 CT patients (mean age = 58 years; 575 (47%) male), 781 (64%) were married and 851 (70%) had metastatic disease. All cancer types were represented, but hematologic cancers (21%) were most common. Within 6 months following CT enrollment, 519 (43%) and 327 (27%) had at least one hospitalization and ED visit, respectively. At any time during their cancer course, 177 (15%) received a palliative care (PC) consult. Controlling for presence of metastatic disease, PC consults correlated with both hospitalizations and ED visits. Having a hematologic cancer and being unmarried correlated with more hospitalizations and ED visits, respectively. Conclusions: Hospitalizations and ED visits occur in a substantial proportion of cancer CT patients. We need to better understand reasons for these high rates of health care utilization, but the correlations with PC consults suggest that CT patients have unique supportive care needs and that PC services are being targeted to a population particularly in need. [Table: see text]

116 Background: PD-1/PD-L1 inhibitors (PD-1/PD-L1i) have revolutionized the treatment of patients with metastatic melanoma. However, little is known about the supportive care needs and use of palliative care (PC) services in these patients. We sought to assess rates and correlates of PC consultation and hospice use among melanoma patients treated with PD-1/PD-L1i. Methods: We retrospectively collected data on all patients with melanoma treated with PD-1/PD-L1i at the Massachusetts General Hospital (MGH) from May 2009 to March 2015. We determined rates of PC consultation and hospice use in this sample. We used logistic regression with purposeful selection of covariates to identify correlates of PC consultation. Results: Of 122 patients treated at MGH (mean age = 61.2 years), most were married (n = 91, 75%) and college educated (n = 66, 54%). Overall, 83 patients (68%) experienced autoimmune toxicities, 22 (18%) had liver metastases, 87 (71%) had cancer progression and 44 (36%) died during our study period. We found that 36 patients (30%) received PC consultation during their treatment course; 9 (25%) of these occurred over 6 months prior to the patients’ death, and 11 (31%) in their last month of life. Among the patients who died, 36 (82%) received hospice services, and 11 (31%) enrolled in hospice during their last week of life. In our regression model, younger age, having cancer progression and presence of liver metastases correlated with increased likelihood of receiving PC consultation. Conclusions: Melanoma patients treated with immunotherapy often receive PC and hospice services late in their treatment course. These data underscore the difficulty understanding when to involve hospice and PC for these patients, and highlight the need for further PC research in this patient population. Additionally, younger age, presence of cancer progression and liver metastases were associated with receiving PC consultation, thus suggesting the need for integrated PC services among certain subpopulations of these patients. [Table: see text]

51 Background: When patients with cancer transition to hospice, the role of oncology clinicians can be unclear. During this time, patients and families often feel abandoned by their oncology team. We assessed the feasibility of a novel intervention to enhance continuity of care between the oncology team and patients and their families after they enroll in hospice. Methods: Family caregivers (FCs) of patients with cancer (lung, head and neck, sarcoma, melanoma, and gynecological) who enrolled in home hospice from 3/2014-6/2015 were eligible to participate in the study. The ECHO intervention consisted of check-in phone calls from the oncology team to the FC (weekly x 4 weeks, biweekly x 4 weeks, then monthly), an optional in-person clinic visit, and a bereavement call. FC questionnaires assessing satisfaction, perceived stress, quality of EOL care, and regret were completed at enrollment, weeks 1 and 2, and then every 4 weeks, as well as at a phone interview post- death. The primary outcome was feasibility, which we defined as > 70% of the FCs receiving > 50% of the phone calls and > 70% of the FCs completing > 50% of the questionnaires. Results: Of 37 eligible FCs, 24 (80%) enrolled, 7 were not reached, and 6 declined. At the time of analysis, 19 were evaluable as 2 patients died before the intervention began, 1 FC withdrew, and 2 interventions were ongoing. Median time from hospice referral to study enrollment was 3 days (range 1-7) and from study enrollment to patient death was 34 days (range, 3-135). Clinicians completed 156 of the expected 172 calls (90.7%) to FCs. FCs completed 75 of the expected 91 (82.4%) questionnaires. 100% of the FCs received > 50% of scheduled phone calls and 79% of the FCs completed > 50% of the questionnaires. In qualitative interviews post-death, FCs indicated that the phone calls were appreciated and helpful. Conclusions: This intervention enhancing communication with the oncology team throughout hospice was well received by FCs. Additionally, the intervention proved feasible, as the majority of phone calls and questionnaires were completed. A randomized study is warranted of the impact of ECHO on the quality of EOL care and FC outcomes including satisfaction, bereavement adjustment, and mood. Clinical trial information: NCT02069769.

76 Background: Patients’ prognostic understanding and coping styles influence their treatment decisions, but how these factors relate to their quality of life (QOL) and mood has not been well described. We sought to examine the associations of patients’ prognostic understanding and coping style with their QOL and mood. Methods: As part of an ongoing trial of early palliative care, we are assessing baseline QOL (Functional Assessment of Cancer Therapy-General), mood (Hospital Anxiety and Depression Scale), coping (Brief Cope) and prognostic understanding in patients within 8 weeks of diagnosis of advanced lung or gastrointestinal (GI) cancer. To determine associations, we used linear and logistic regression, controlling for patients’ age, sex, cancer type and marital status. Results: Of 300 participants (mean age = 64.7 years; 138 (46%) female), 132 (44%) had GI cancer and 168 (56%) had lung cancer. Using cutoff score > 7 for the HADS, 61 (20%) and 85 (28%) reported depression and anxiety. 138 (49%) reported their prognosis as terminal. A terminal perception of prognosis was associated with lower QOL and higher rates of anxiety. Emotional support, acceptance, and active coping styles were associated with better QOL and mood. Denial and self blame were associated with worse QOL and mood. Conclusions: These data demonstrate that acknowledging a terminal prognosis may be associated with greater physical and psychological distress, or conversely, patients with worse QOL and mood may better appreciate the gravity of their illness. Certain coping styles (self blame and denial) are associated with lower QOL and higher distress. Understanding the relationships among patients’ prognostic awareness, coping styles, QOL and mood will allow us to develop more effective supportive care interventions. Clinical trial information: NCT01401907. [Table: see text]