Journal of Oncology / 2014 / Article / Tab 2

Clinical Study

Development of a Score Predicting Survival after Palliative Reirradiation

Table 2

Prognostic factors for survival. All baseline variables shown in Table 1 were analyzed (univariate, log-rank test). Those with value <0.1 were carried forward to multivariate Cox regression analysis and are shown here.

CharacteristicMedian survival (months)value
Univariate1Multivariate

Karnofsky PS
 90–10018.30.00010.0001
 70–809.4
 ≤602.1
Known brain metastases
 No 9.70.008n.s.
 Yes3.6
Known liver metastases
 No9.70.0370.039
 Yes2.8
Pleural effusion
 No9.40.0070.039
 Yes1.3
Number of metastatic sites
 Max. 29.70.054n.s.
 3 or more2.8
Progressive disease outside RT target volume
 No12.60.033n.s.
 Yes5.5
Use of opioid analgetics
 No24.50.02n.s.
 Yes5.2
Use of steroids
 No12.20.0020.015
 Yes3.6
Serum albumin
 Low9.70.001n.s.
 Normal2.8
Serum alkaline phosphatase
 Normal15.10.027n.s.
 Elevated4.1
Serum creatinine
 Low1.60.0001n.s.
 Normal9.7
 Elevated15.1
Serum C-reactive protein
 Normal18.30.0001n.s.
 Elevated but less than 30 mg/L12.6
 Elevated 30–60 mg/L5.3
 Elevated >60 mg/L2.6
Thrombocyte count
 Low12.70.038n.s.
 Normal9.7
 High4.0
Number of abnormal blood tests2
 Max. 112.7
 25.80.008n.s.
 3 or more3.0
Smoking status
 Current smoker4.30.063n.s.
 No9.7
Time from first cancer diagnosis
 Shorter than median (47 months)5.30.089n.s.
 Longer than median9.7

RT: radiotherapy; PS: performance status.
1If more than 2 groups, value from log-rank test pooled over all strata.
2All tests shown in Table 1 were considered.
Significance levels were not corrected for multiple tests.