Radiation for Pancreatic Adenocarcinoma: An End-of-Life Treatment? Assessing Appropriate Radiation Therapy Utilization in the National Cancer Data Base (NCDB) International Journal of Radiation Oncology, Biology, Physics EPOSTER ABSTRACT| VOLUME 96, ISSUE 2, SUPPLEMENT , S195, OCTOBER 01, 2016
Date
10/01/2016Abstract
Purpose/Objective(s)
The 5-year overall survival of patients diagnosed with pancreatic adenocarcinoma (PAC) is a dismal 7%. Guidance from ASTRO’s Choosing Wisely campaign suggests that treatment benefits and burdens should be weighed carefully in the face of a limited prognosis. As yet, no distinct recommendations exist to guide effective utilization of radiotherapy (RT) in patients with advanced PAC. We aim to describe the current patterns of RT utilization in PAC patients.
Materials/Methods
The National Cancer Data Base (NCDB) was queried for patients with pancreatic adenocarcinoma receiving any form of RT between 1998 and 2006. Receipt of late RT (RT in the last 60 days of life), RT prior to the last 60 days of life, fractionation schedules, and patient and facility demographics were described. Univariate (UV) analysis of covariates associated with late RT was performed using ANOVA or chi-square test. UV and multivariate (MV) logistic regression were performed to determine predictors of late RT.
Results
A total of 147,023 pancreatic adenocarcinoma patients were identified from NCDB. Of the 20% (n = 29,849) who received RT and had death status recorded, 53% (15,741) died within 60 days of receiving RT. Eighty-three percent (23,463/28,421) of all patients with fractionation data available received >15 fractions of RT, including 47% (1,774/3,770) of all stage IV patients, 86% (6,144/7,111) of all stage III patients, 87% (6,318/7,271) of all stage II patients, and 89% (2,972/3,350) of all stage I patients. Fifty-two percent (271/517) of patients receiving RT to the bone received >15 fractions, as did 10% of patients receiving brain RT and 8% of patients receiving spine RT. On MV analysis, significant factors associated with receipt of late RT included year of treatment prior to 2004 (HR = 4.10, 95% CI = 3.89-4.32; P < 0.001); stage I-III disease (HR = 2.52, 95% CI = 2.40-2.64; P < 0.001); age <75 years (HR = 2.02, 95% CI = 1.91-2.13; P < 0.001); and treatment at an academic center (HR = 1.12, 95% CI = 1.12-1.17; P < 0.001).
Conclusion
In this large, national cohort, an unanticipated 53% of all patients receiving RT for PAC died within 60 days and the majority of these spent greater than three weeks receiving daily RT. A significant portion of patients with advanced stage disease or receiving treatment to palliative sites received fractionation courses far longer than established guidelines. This novel data should bring into question appropriate utilization of RT for a disease with a very poor prognosis and guide further exploration into shorter courses of RT for patients with PAC. Prognostic models should be developed to avoid lengthy treatment courses with limited benefits and potential toxicities near end-of-life.
Author List
L.E. Colbert K.V. Dharmarajan J. Switchenko C.M. Taniguchi W.A. Hall T.W. Gillespie D.A. Kooby J. Lipscomb J.C. LandryAuthor
William Adrian Hall MD Chair, Professor in the Radiation Oncology department at Medical College of WisconsinView Online
