Predicting Time of Remaining Life on Treatment During Palliative Radiotherapy: A Stratified Analysis of a Novel Palliative Use Ratio.

PURPOSE/OBJECTIVE(S) To investigate factors associated with high proportion of remaining life spent receiving palliative radiotherapy (PRT) for three commonly utilized PRT regimens using a novel metric, the Palliative Use Ratio (PUR). MATERIALS/METHODS We identified consecutive patients receiving PRT with 8 Gy in a single fraction (n = 207), 20 Gy in 5 fractions (n = 152), and 30 Gy in 10 fractions (n = 651) at the discretion of the treating clinician between 2014-2018. The number of days between PRT start and finish was calculated. This value was then divided by overall survival (OS, in days from start of PRT) to generate the PUR, which is summarized using median (interquartile range, IQR) and compared across various clinical factors using the Kruskal-Wallis test. Factors identified with a significance level P < 0.01 were included in a novel PUR Score (PURS). Each factor, if present, equals 1 point. This score was used to categorize and predict for high PUR across each fractionation scheme. RESULTS A total of 1010 PRT courses were delivered. Baseline factors associated with fractionation included ECOG, PRT indication, prior PRT, chemotherapy, and treated site. The fractionation scheme chosen was ultimately associated with OS. Factors associated with increased PUR were included in the scoring system and stratified across 1, 5, and 10 fraction regimens, as outlined in the table below. For patients treated with 8 Gy in 1 fraction, low (0-1) v. high (2-5) PUR Score was associated with a median PUR of 0.01 v. 0.03 (P < 0.001). For those treated with 20 Gy in 5 fractions, low (0-1) v. high (2) PUR Score was associated with a PUR 0.05 v. 0.15 (P < 0.001). A three-tiered PUR Score categorization (low, 0-2; medium, 3-4; high, 5-6) predicted increased PUR in patients treated with 10 fractions: 0.06 v. 0.13 v. 0.23, respectively (P < 0.001). A high PUR Score predicted for an increased ratio of time on treatment to remaining life: 3% for 1 fraction, 15% for 5 fractions, and 23% for 10 fraction regimens. CONCLUSION We describe a novel metric for evaluating the utility of three common PRT regimens that may assist with clinical decision-making. Single-fraction regimens are associated with low PUR. When considering 5 or 10 fraction regimens, the PUR Score identifies patients who may benefit from shorter courses of PRT, as those with a high PUR score spent a substantial proportion of their remaining life on treatment. Prospective, external validation is warranted.