Does use of total parenteral nutrition in patients with metastatic gastrointestinal cancer impact hospice length of stay?

58

Background: The use of total parenteral nutrition (TPN) in metastatic cancer patients is controversial. The impact of TPN use on length of hospice stay in advanced cancer patients is unknown. We hypothesize that patients with metastatic gastrointestinal cancers who receive TPN have a shorter median length of hospice stay as compared to the median hospice length of stay for cancer patients in the United States in 2014 (18 days). The primary objective of our retrospective, single-institution study is to determine the median hospice length of stay in patients with metastatic gastrointestinal cancers who received TPN. Methods: Records of all adult patients with metastatic gastrointestinal cancers who received TPN at The Lifespan Cancer Institute from 2005 through 2014 were reviewed. The primary outcome was median hospice length of stay. Data analysis was conducted using Stata (Version 15.0, StataCorp, College station, Texas). Results: Seventy-nine patients were identified as having received TPN for metastatic gastrointestinal cancer. Forty-eight patients had documented referrals to hospice and 40 patients had assessable durations of hospice admission. Hospice length of stay ranged from one to 196 days with a median of 9.5 days, mean of 24.3 days, and interquartile range of 5.5 to 54 days. Conclusions: Hospice care has been shown to improve quality of life for patients and caregivers. In our retrospective, single-institution study of patients with metastatic gastrointestinal cancer who received TPN, the median hospice length of stay was shorter than the national median length of hospice stay for cancer patients in 2014 (9.5 days versus 18 days). This should be taken into consideration when weighing the risks and benefits of initiating TPN in metastatic gastrointestinal cancer patients. [Table: see text]

Strategies to embed palliative care into a culture of cancer care

e21678

Background: In 2012, the 3 hospital Lifespan Health System launched a palliative care initiative. The hospitals and medical oncologists knew this was critical for patient centered and value-based cancer care, but recognized many barriers: physician practice patterns, lack of dedicated resources and systems, patient and family education gaps, and limited return on investment in the current environment. A multi-level inpatient and outpatient strategy was implemented and tracked over four years. Methods: External benchmarking data from a Medicare claims analysis of Vizient (academic health system consortium) member organizations and from ASCO QOPI data were used in the analysis. Internal data analysis included a study on symptom management for lung cancer patients, hospital reports on palliative care service utilization, ED visits and hospital admission trends for cancer patients. Multi-level interventions were employed: hospital investment in staff and systems, partnership with a community-based hospice and palliative care provider, a medical oncology physician champion with Board certification in palliative care, a palliative care inpatient consult service and daily ICU rounds, an oncology medical home, medical oncologist Saturday hours, electronic prompts for consults, and a cancer call triage center. Results: A Medicare claims analysis for 2012 to 2014 on cancer decedents with ICU stays in the last 30 days in the Vizient national study of health systems showed that Lifespan was at the 11th percentile, making them the 4th lowest (days in ICU) in performance (pre/post data requested). QOPI data on appropriate referrals to hospice or palliative care prior to death improved from 58% in 2010, which was below the QOPI benchmark of 61%, to 94% in 2016 which is above the QOPI benchmark of 74%. Other QOPI and hospital data will be included in the presentation. Conclusions: Palliative care, a crucial tool for the delivery of future cancer care, is challenging to implement effectively. This study shows that a hospital/medical oncology partnership can drive change to embed palliative care into the culture of cancer care and these strategies offer a roadmap for others to follow as they strive to offer patient centered and value-based cancer care.

Development of a screening tool to assess prostate cancer health literacy

127

Background: Poor comprehension of prostate cancer (PCa) related terms can create barriers to informed discussions on screening, treatment, and measuring outcomes, and contribute to disparities in African American (AA) men. We developed a screening tool to assess for low PCa-related health literacy. Methods: We assessed PCa-related literacy in a sample of 189 AA men, age >40 years from diverse socioeconomic status (SES) using a 27-item scale derived comprehension of commonly used terms for urinary, bowel and sexual function. Using item-response models we examined differential item functioning by education. We developed rapid screening tools based on understanding of 1 or 2 words to predict overall comprehension. Receiver operating characteristic curves assessed the sensitivity and specificity for individuals understanding less than a pre-specified threshold of 70% on the overall scale, defined as “low literacy.” Results are being tested in an independent sample of 110 AA men. Results: The 27-item scale had good internal reliability (Cronbach alpha = 0.93). 47% of the sample met criteria for low literacy. Lower education groups had relatively poor comprehension of sexual function terms compared to higher education groups. 1-item scales using comprehension of the term “rectal urgency” had a sensitivity of 95% for identifying low literacy, “erection” had a specificity of 98%, and “vaginal intercourse” had a sensitivity of 91% and specificity of 81%. Combining “vaginal intercourse” and “rectal urgency” yielded a 2-item scale with strong characteristics (sensitivity 88%, specificity 89%), as did combining “vaginal intercourse” and “erection” (sensitivity 94%, specificity 81%). Conclusions: Rapid screening tools assessing PCa-related literacy performed well in a community sample of AA men with varied SES. Providers can use these tools to identify those at risk of poor comprehension to tailor outcome measurement and shared decision making. [Table: see text]

Outpatient palliative care encounters in stage IV lung cancer care: An institutional review

128

Background: Palliative Care (PC) is becoming increasingly integrated into standard oncologic care (SC). Previous research suggests that patients receiving PC report better quality of life, and may have prolonged survival. This study evaluates the effect of PC integration in patients diagnosed with stage IV non-small cell lung cancer (NSCLC) at a single institution. Methods: All patients diagnosed with Stage IV NSCLC between January 2010 and January 2013 were considered for inclusion and retrospective analysis of their care. Charts were reviewed to identify patients who received outpatient PC with a licensed PC physician in addition to SC. There were no guidelines regarding the nature of the PC intervention. Retrospective analyses of multiple factors were assessed, including: receipt of chemotherapy and/or radiotherapy, utilization of emergency and sick visits, frequency and timing of hospice referral, and duration of hospice utilization. Overall survival was also assessed. Results: 136 patients fulfilled study inclusion criteria. 29 patients received PC in addition to SC, and 107 received SC alone. No statistically significant difference was noted between the groups with respect to age, sex, lines of chemotherapy administered, number of emergency department visits, or number of clinic sick visits. Hospice was offered more frequently in the PC group; however, there was no difference in the amount of time spent on hospice, and no difference in overall survival. There was a trend towards longer survival in the PC group (220 days vs. 254 days). Patients seen in a multidisciplinary clinic were significantly more likely to receive a PC evaluation (RR 1.28 CI 1.073-1.52, p < 0.006). Conclusions: This retrospective study examines how PC is integrated in actual clinical models. Multidisciplinary clinic patients were more likely to receive PC after controlling for comorbidities. There was no significant difference between PC and SC group outcomes. Although this study is small, it demonstrates common practice patterns, and identifies the need to identify the components of the PC encounter that are important in order to maximize the potential benefits of PC interventions.

Perspectives on palliative care in a multisite oncology practice

52

Background: The Institute of Medicine reported that many cancer patients do not receive palliative care to manage symptoms. It is possible that practitioners do not identify palliative care as an important component of high-quality cancer care, or do not know how to most efficiently utilize available palliative care consultative services. Methods: An anonymous electronic survey was sent to physicians, nurses, nurse practitioners, physician assistants, and social workers (n=99) in our multi-site, single-institution Cancer Center. Results: Sixty-five responses were received (66% response rate). Eighty-three percent of respondents reported working primarily in the outpatient setting. Fifty-nine percent reported their patients ‘rarely use the ER for pain management’ while 16% reported their patients ‘frequently require ER visits for pain management’. Ninety-two percent considered palliative care ‘an integral part of a multidisciplinary team’ while 6% reported palliative care consultation to be ‘cumbersome to consult and coordinate with’. The most common reason for consultation was end-of-life discussions (38%) followed by chronic pain management (33%). Seventy-seven percent reported consulting between 1 and 5 times per month, 14% between 6-10 times per month, and 5% greater than 11 times per month. Thirteen percent reported that they were able to manage patient symptoms adequately themselves. The average rating for convenience of consulting palliative care was 3.8/5. Fifty-eight percent reported the most effective means of communication with palliative care consultants was through the medical record, whereas 42% reported that most effective communication took place in person. Sixty percent reported a preference for palliative care practitioners from within the division of hematology/oncology. Conclusions: Our survey suggests that the majority of oncology practitioners value palliative care consultation and are willing to incorporate palliative care services into patient management. It is possible that practitioners overestimate their utilization of palliative care services and that optimizing the convenience of consultation and communication would result in better integration of palliative care for cancer patients.

Phase I study of ridaforolimus with cetuximab for patients with advanced non-small cell lung cancer (NSCLC), colorectal cancer, and head and neck cancer

8075

Background: mTOR inhibition may overcome PI3K/AKT pathway mediated resistance to anti-EGFR therapy. We performed a phase I study to determine the dose-limiting toxicity (DLT) of ridaforolimus, an investigational oral mTOR inhibitor, in combination with the anti-EGFR antibody cetuximab. Methods: Patients with advanced NSCLC, colorectal cancer, and head and neck cancer that progressed after at least 1 prior regimen for metastatic disease were eligible. ECOG performance status 0-1. Patients with previously treated brain metastases that were stable for >3 months were eligible. Wild-type K-RAS was required in colon cancer. All patients received cetuximab 400 mg/m2 week 1 followed by 250 mg/m2 weekly. Three dose levels of ridaforolimus were planned: 20mg, 30mg, and 40mg daily, 5 days each week, on a 28-day cycle. Results: 12 patients were entered with NSCLC (n=7), colon cancer (n=4), and head and neck cancer (n=1). The median age was 58 (42-69). The median number of prior regimens for metastatic disease, by disease type, was NSCLC (n=3), colorectal (n=4), head & neck (n=4). Three patients completed the first dose level without DLT. Two of 3 patients at dose level 2 had dose-limiting mucositis. The first dose level was then expanded with six additional patients with NSCLC without any further dose-limiting toxicities. The recommended phase II dose of ridaforolimus is 20 mg daily, 5 days a week, in combination with cetuximab. Response and prolonged stable disease was demonstrated in NSCLC. Conclusions: The DLT of the combination of ridaforolimus and cetuximab is mucositis. The activity observed in heavily pretreated patients with NSCLC suggests that the combination of an mTOR inhibitor with an EGFR antibody merits further investigation in NSCLC. Clinical trial information: NCT01212627.

Lenalidomide for second-line treatment of advanced hepatocellular cancer (HCC): A Brown University Oncology Group phase II study

4098

Background: Lenalidomide inhibits fibroblast growth factor (FGF), vascular endothelial growth factor (VEGF) and multiple tumor growth pathways. There is no standard of care for patients who progress after sorafenib. Therefore, we performed a phase II study to determine the activity of lenalidomide in second-line HCC therapy. Methods: Patients with advanced HCC who progressed on or were intolerant to sorafenib were eligible. Prior chemoembolization, RFA, or surgery were allowed. Eligibility criteria also included bilirubin <4 mg/dL, AST and ALT <5 times upper limit of normal, ECOG performance status 0-2, platelet count >60,000/mm3, absolute neutrophil count >1000/mm3, and creatinine <2mg/dL. Patients were treated with lenalidomide 25mg orally days 1-21 of a 28 day cycle until disease progression or unacceptable toxicities. The planned original sample size was 25 patients but when early activity was demonstrated the study was expanded to 40 patients. Results: The study has completed accrual of 40 patients. The median age was 60.5 years (17-88 years). Nineteen patients were Child-Pugh A, 16 patients were B, and 5 patients were C. Twenty four patients had extrahepatic disease. Preliminary data is available on the first 37 patients. One patient had grade 4 neutropenia. Grade 3 toxicities included ANC (n=2), fatigue (n=4), rash (n=2), arthritis (n=1), diarrhea (n=1), dehydration (n=2). One patient developed variceal bleeding which precipitated encephalopathy and death. Of the 32 patients with elevated baseline AFP, nine (28%) had a >50% reduction including one patient with a reduction in AFP from 56,900ng/ml to 5 ng/ml. Six of the first 37 patients (16%) had a radiographic partial response. Two patients achieved a complete response and have not progressed at 36 and 32 months. Conclusions: Lenalidomide can be administered to patients with advanced HCC and significant hepatic dysfunction. Promising, and in a small percentage of patients, dramatic and durable activity has been demonstrated. Investigations are underway to explore the mechanism of action of lenalidomide in HCC.