The impact of early palliative care on the quality of life of patients with advanced pancreatic cancer: The IMPERATIVE study

12116

Background: Pancreatic cancer (PDAC) is an aggressive, deadly disease. Chemotherapy (CT) can improve survival by months, but symptom burden is heavy and quality of life (QOL) is poor. Early palliative care (EPC) alongside standard oncologic care improves QOL and survival in other types of cancer; however, the impact on QOL and symptom burden in advanced PDAC is not known. The primary objective of this study was to test for improvement in QOL between baseline (BL) and 16 weeks (wks) among patients receiving EPC. A secondary objective was to test for decreased symptom burden between BL and 16 wk. Methods: In this prospective case-crossover study, patients >18 years with advanced PDAC received EPC provided by a subspecialist palliative care physician and advanced practice nurse plus standard oncologic care. Ambulatory EPC visits occurred every 2 wks for the first month, then every 4 wks until wk 16, and then as needed. The Functional Assessment of Cancer Therapy – hepatobiliary (FACT-hep) and Edmonton Symptom Assessment System (ESAS) questionnaires were completed at enrollment and every 4 wks until wk 16. Least square means and 95% confidence intervals were computed. A generalized linear mixed model was used to test for statistically significant change in scores between BL and wk 16. A sample size of 20 patients provides 80% power after controlling for covariates; 40 patients were enrolled to account for anticipated attrition and missing data. Results: Of 40 patients, 25 (62.5%) were male, 28 (70%) had metastatic disease, 31 (77.5%) had an ECOG performance status of 0-1, 17 (42.5%) had a body mass index (BMI) >25, 35 (89.7%) had an elevated CA19-9 and 31 (77.5%) received CT. Median age was 70.2 (range 63.0-77.5). BL and wk 16 questionnaires were completed by 100% and 70% of patients, respectively. The mean FACT-hep score at BL was 118.8, compared to 125.7 at wk 16, for a mean change of 6.89, [95%CI (-1.69-15.6); p = 0.11]. The mean change from BL to wk 16 for FACT-hep was statistically significant in patients receiving CT, 10.1 [95%CI (0.32-19.8); p = 0.04], patients with metastatic disease, 14.7 [95%CI (5.30-24.1); p = 0.0030] and patients with a BMI >25, 12.5 [95%CI (1.29-23.7); p = 0.03]. The mean ESAS total symptom score at BL was 25.3, compared to 22.7 at wk 16 (p = 0.28). In those with metastatic disease the mean change was statistically significant, -5.73 [95%CI (-11.21 to -0.24); p = 0.04]. Conclusions: EPC resulted in improved QOL in pts with PDAC receiving CT and those with a BMI >25, and improved QOL and symptom burden in patients with metastatic disease. Given minimal attrition and high rates of questionnaire completion, our sample size was robust, resulting in strong power. Providing palliative care alongside standard oncologic care results in clinically meaningful improvements. Access to palliative care, shortly after diagnosis, should be available for patients with advanced PDAC. Clinical trial information: NCT03837132.

The impact of early palliative care on the quality of life of patients with advanced pancreatic cancer: The IMPERATIVE study

TPS777

Background: Pancreatic cancer is lethal. Chemotherapy can improve survival by months; however, many patients experience an overwhelming burden of cancer-associated symptoms and poor quality of life (QOL). Early palliative care (EPC) alongside standard oncologic care results in improved QOL and survival in patients with lung cancer. Although international guidelines recommend EPC for patients with advanced pancreatic cancer (PANC), the benefit is not known. Objectives: The primary objective is to test for change in QOL between baseline (BL) and 16 weeks (wk). Secondary objectives are to test for change between BL and 16 wk in (a) symptom control; and (b) depression and anxiety. Methods: This prospective case-crossover study of patients with PANC provides EPC plus standard oncologic care. Primary oncology clinics refer patients to an EPC team led by a palliative care physician and a clinical nurse specialist. BL questionnaires are completed prior to initial EPC assessment, then every 4 wk until wk 16. EPC visits are every 2 wk for the first month, every 4 wk until wk 16, and then as needed. QOL, symptom control, anxiety and depression are measured using the FACT-Hep tool, ESAS-r, HADS and PHQ-9, respectively. A generalized linear model will test for statistically significant change in scores between BL and 16 wk; chemotherapy (yes/no) is included as a confounding covariate; model fit will be assessed. A sample size of 20 patients provides 80% power after controlling for covariate effects. 40 patients will be enrolled to account for missing data. To date, 28 patients have enrolled and 17 have completed the intervention. Significance: The benefit of EPC for patients with PANC is not known, however, EPC is increasingly recognized internationally by patients and stakeholders as a critical intervention which may improve both QOL and satisfaction with care. The Canadian Partnership Against Cancer’s report on the patient experience states “the best possible patient experience means all people with cancer have equitable access to high-quality person-centered palliative care”. This study offers access to EPC and provides an environment in which the benefit of an integrated approach is evaluated.

Impact of a breast cancer (BC) polygenic risk score (PRS) on the decision to take preventive endocrine therapy (ET): The Genetic Risk Estimate (GENRE) trial

1501

Background: Despite BC risk reduction of 50-65% by preventive endocrine therapy (ET), very few at-risk women choose to take them. A woman’s perceived BC risk correlates with uptake of ET. A PRS comprised of 77 BC genetic susceptibility loci (Single Nucleotide Polymorphisms (SNP) improves the accuracy of risk prediction for BC. We examined the impact of the addition of individualized PRS BC risk prediction to standard risk calculator estimates on intent to take BC prevention medication. Methods: Eligible women had ≥5% 10 yr BC Tyrer-Cuzick risk (IBIS) or 5 year Gail score ≥3%, with no history of BC or hereditary BC syndrome. Standard BC risk estimates (IBIS or Gail) were incorporated into the counselling on BC preventive ET. A self-reported questionnaire at baseline quantified intention to take ET and explored factors associated with this decision. Blood samples were obtained and genotyped for 77 SNPs, individualized PRS were calculated then incorporated into IBIS and Gail predictions for 5 yr, 10 yr, & lifetime BC risk. At a second visit, PRS risk & prevention recommendations were revisited. Post visit questionnaires assessed change in intent to take ET. Multivariable linear regression was performed to assess impact of baseline variables on change in intent to take medication. Results: From 2016 to 2017, 151 women in Canada & USA were enrolled, median age: 56.1 (range 36-76.4), 35.6% were premenopausal, 98.7% were Caucasian. Median 5yr, 10yr, & lifetime IBIS risk estimates were 3.8% (2.0-11.5), 7.9% (5.0-23.1), and 25.3% (5.5 to 92.2). PRS increased BC risk estimates in 84 (55.6%) and reduced BC risk estimates in 67 (44%) women. After PRS risk counselling, intention to take ET significantly changed (p<0.001): 41.9% of those with increased PRS were more inclined, and 46.7% of women with decreased PRS were less inclined to take ET. On multivariable regression, increase in PRS (p<0.0001) and less concern about ET side effects (p<0.0001) were associated with greater intent to take ET. Conclusions: In high risk women, PRS significantly changed BC risk estimates & intent to take preventive ET. Further assessments of the impact of PRS scores on compliance with ET are warranted. Clinical trial information: NCT02517593.

Interventions to improve oral chemotherapy safety and quality: A systematic and grey literature review

97

Background: With the growing use of oral chemotherapy, there is an urgent need to develop safe and effective systems to administer and manage these agents. A comprehensive synthesis of literature on oral chemotherapy care delivery programs to which clinicians can look for best practices is lacking. Methods: We performed a systematic review of PubMed, EMBASE and CINAHL from 1/1995- 5/2016 and the grey literature to identify publications describing oral chemotherapy care delivery programs. Our population of interest was cancer patients of all ages prescribed cytotoxic or targeted anti-cancer oral agents. Interventions could address any part of the oral chemotherapy delivery process from prescribing through disposal but had to report outcomes (adherence and/or safety or toxicity) in relation to a control group. Results: From 7,984 abstracts in the peer-reviewed and 9 from the grey literature, 16 studies met inclusion criteria (7 of these randomized) with 3,612 patients represented. Interventions focused on prescribing (n = 1), preparation/dispensing (n = 2), education (n = 11), administration (n = 5), monitoring (n = 14), and storage/disposal (n = 1). Of the 10 articles with adherence as an outcome, four different measurement methods were used. Many articles lacked formal statistical testing. In the 6 studies with statistically significant improvement in outcomes, 3 utilized nursing phone calls to patients within the first few days of treatment initiation, which resulted in less toxicity (n = 2) or better adherence (n = 1). None of the four studies that evaluated eHealth strategies to increase patient to care team contact demonstrated a statistically significant improvement in outcomes. Conclusions: Limitations in study design impair our ability to draw definitive conclusions on best practices for oral chemotherapy care delivery. A framework for conducting research in this area that defines the processes of oral chemotherapy delivery and standardizes outcomes of success is needed to address this gap. Existing studies suggest that interventions focusing on education and remote phone-based monitoring of patients at therapy initiation may decrease toxicity, and possibly improve adherence.

Resection of isolated metastases to the pancreas: An institutional retrospective review

365

Background: Resection of isolated metastases in good performance status patient (pts) has increasingly become an accepted standard of care. While resection of hepatic and pulmonary metastases from selected cancers such as CRC and RCC have supportive evidence, the safety and effectiveness of pancreatic resections for metastases from these same tumour sites is not as well established. We aim to identify the optimal assessment and selection criteria which would best predict a favourable outcome from pancreatic metastatectomy. Methods: A prospective pathology database was used to identify pts who underwent partial or total pancreatectomy for solid tumour metastases to the pancreas between Sept 2000 to May 2013. Demographic, histologic, surgical and survival data was collected and analyzed using descriptive statistics. Results: 19 pts were identified and included. Follow-up time ranged from 5 months to 8 years. Most common primary sites were RCC (42%,) CRC (11%) and GIST (11%). Favourable outcome, defined as an event-free survival greater than 12 mos from pancreatectomy, was observed in 63% (12/19) and was associated with a higher proportion of RCC (50% vs 14%), pancreas as M1 presentation (58% vs 43%), longer time from presentation to resection (median 5mos vs 2mos), multidisciplinary conference (MDC) review (33% vs 14%) and preop PET (42% vs 14%). No differences were observed by age or gender. Median postoperative length of stay was 22.1 days. Conclusions: Resection of pancreatic metastases remains a rare indication. Given the morbidity and extended LOS associated with pancreatectomy, careful patient selection is required. In this small retrospective series, favourable outcomes were more frequently associated with RCC primary, delayed resection, pre-op MDC and PET staging. Multicentre surgical registries would enhance future analysis and better guide patient selection.