PV QA 3 - Poster Viewing Q&A 3
TU_29_3028 - Determination of Physician Effectiveness in Adjusting Palliative Radiation Fractionation for Patients Near the End of Life
Tuesday, October 23
1:00 PM - 2:30 PM
Location: Innovation Hub, Exhibit Hall 3
Determination of Physician Effectiveness in Adjusting Palliative Radiation Fractionation for Patients Near the End of Life
V. Cassidy1, H. Perlow1, A. W. Awerbuch1, D. Kwon2, J. Quintana1, J. Griggs1, S. Ciraula1, S. Alford1, R. Yechieli3, and S. Samuels3; 1University of Miami Miller School of Medicine, Miami, FL, 2Department of Statistics, University of Miami Sylvester Comprehensive Cancer Center, Miami, FL, 3Department of Radiation Oncology, University of Miami / Sylvester Comprehensive Cancer Center, Miami, FL
Purpose/Objective(s): Many patients with metastatic disease receiving palliative radiation die shortly after completing prolonged radiotherapy (RT) schedules. This suggests that radiation oncologists (ROs) are not able to accurately predict patient survival and may be overly optimistic about patient prognosis. Established guidelines stipulate that patients with limited life expectancy are more appropriately treated with fewer fractions to reduce the treatment burden at the end of life, but adherence to these guidelines is variable. The purpose of this study is to determine whether palliative RT fractionation schedules correlate with time until death (TUD) after RT and evaluate demographic and patient related factors to predict prognosis in metastatic patients.
Materials/Methods: This retrospective study included patients treated with palliative RT for metastatic cancer to the brain, spine, and bone between January 1, 2016 and June 30, 2016 at either a safety-net hospital (SNH) or adjacent private academic hospital (PAH). Variables analyzed included treatment location, gender, age, race, language preference, socioeconomic score, comorbidity, Karnofsky Performance Status (KPS), and intended number of fractions (NOF). NOF was treated as a binary variable with patients categorized as receiving either 5 or fewer fractions or greater than 5 fractions. Univariable analysis (UVA) and multivariable analysis (MVA) were performed by using the Cox proportional hazards regression model. All p-values were two-sided. Hazard ratios (HRs) and corresponding 95% confidence intervals and p-values were estimated.
Results: 145 patients were eligible for inclusion in this study. 34.5% (50) patients were treated for brain metastases and 65.5% (95) patients were treated for metastases to the bone or spine. 80.0% of patients were treated at the PAH, and 20.0% of patients were treated at the SNH. 46.9% of patients identified as Hispanic, 32.4% of patients identified as non-Hispanic White, and 18.6% of patients identified as non-Hispanic Black. 10% of brain patients and 17.9 of spine/bone patients had a KPS ≤60. 2% of brain patients and 17.9% of spine/bone patients received 5 or fewer fractions. Upon MVA, poor KPS (≤60) vs. good KPS (≥90) was associated with shorter TUD in both brain (HR 8.72, p = 0.02) and spine/bone (HR 7.25, p < 0.01) palliative radiation patients. Planned number of fractions had no relationship with TUD in either disease site.
Conclusion: These results are consistent with other studies that link poor functional status with a shorter TUD. There was no correlation between fractionation schedule and TUD, which is consistent with ROs not properly tailoring their radiation schedules for patients’ poor prognoses. To improve this, ROs should elevate the importance of performance status when determining fractionation schedules and be more conscious of not giving prolonged courses of treatment to patients with limited life expectancy. Prospective trials to evaluate and improve this metric are necessary.
Author Disclosure: V. Cassidy: None. H. Perlow: None. A.W. Awerbuch: None. D. Kwon: None. J. Quintana: None.