A257 PANCREATIC CANCER TREATMENT AND END OF LIFE OUTCOMES: A POPULATION BASED COHORT STUDY

Abstract

BackgroundPatients with pancreatic cancer face challenging decisions regarding treatment choices following their diagnosis and often lack data on end-of-life (EOL) outcomes. Without the available information, older patients may be undertreated, dying earlier than they would have with treatment, while others may be overtreated and exposed to aggressive measures with harmful side effects.AimsTo describe survival and EOL outcomes among pancreatic cancer patients based on index cancer treatment, disease stage, and patient characteristics.MethodsWe conducted a population based cohort study in Ontario, Canada of patients who died from April 2010 to December 2017 and were diagnosed with pancreatic cancer prior to death. We used administrative databases to collect data on demographics, baseline health status, treatments, and outcomes. The primary exposure was index cancer treatment (no treatment, radiation, chemotherapy alone, surgery alone, and surgery with chemotherapy). The primary outcomes were mortality, health care encounters per 30 days in the last six months of life, and palliative care visits per 30 days within the last six months of life. Secondary outcomes were location of death (institution vs. community), hospitalization within the last 30 days of life, and receipt of chemotherapy within the last 30 days of life. We estimated the association between the exposure and outcomes using multivariable models, adjusting for demographics, comorbidities, and cancer stage. Hazard ratios, adjusted mean differences, and odds ratios were reported with 95% confidence intervals.ResultsOur cohort included 9950 adults with a median age at diagnosis of 78. 56% received no index treatment, 5% underwent radiation, 27% underwent chemotherapy alone, 7% underwent surgery alone, and 6% underwent surgery and chemotherapy. In the multivariable regression (Table and Figure), radiation, chemotherapy alone, surgery alone, and surgery with chemotherapy were all associated with decreased mortality and fewer healthcare encounters. All groups except radiation were associated with fewer palliative care visits. All treatment groups were associated with lower odds of institutional death and hospitalization within the last 30 days of life, and higher odds of chemotherapy within the last 30 days of life.ConclusionsOur data, the first to provide EOL outcome estimates based on index cancer treatment, can help patients make initial treatment decisions after a diagnosis of pancreatic cancer.Multivariable regression analyses predicting primary and secondary outcomesTreatment groupMortalityHealthcare encounters per 30 daysPalliative care visits per 30 daysInstitutional deathHospitalization within last 30 days of lifeChemotherapy within last 30 days of lifeNo treatmentReferentReferentReferentReferentReferentReferentRadiation0.63 (0.57 to 0.70)-3.64 (-4.71 to -2.58)-0.52 (-1.07 to 0.03)0.79 (0.66 to 0.97)0.66 (0.54 to 0.80)4.13 (2.06 to 8.27)Chemotherapy alone0.43 (0.41 to 0.45)-6.35 (-6.92 to -5.78)-1.57 (-1.87 to -1.28)0.72 (0.65 to 0.79)0.72 (0.65 to 0.80)57.44 (39.71 to 83.08)Surgery alone0.32 (0.29 to 0.34)-6.91 (-7.82 to -6.00)-1.65 (-2.13 to -1.18)0.86 (0.73 to 1.02)0.63 (0.53 to 0.74)10.77 (6.66 to 17.43)Surgery and chemotherapy0.23 (0.21 to 0.26)-6.74 (-7.75 to -5.74)-1.67 (-2.19 to -1.14)0.70 (0.59 to 0.85)0.63 (0.52 to 0.75)21.81 (13.94 to 34.12) Association between index cancer treatment and primary outcomes.Funding AgenciesCIHR