Page 107 - Predicting survival in patients with spinal bone metastasesL

SINS Table 3. Evaluation of SINS accuracy. PPV: positive predictive value. NPV: negative predictive value.
Sensitivity
Specificity
PPV
NPV
Observer 1 Observer 2 Observer 3 Observer 4
69% 48% 18% 90% 35% 27% 7% 71% 35% 18% 6% 63% 41% 25% 8% 71%
Univariate Fine and Gray analysis showed that the final SINS classification, total radiotherapy dose, as well as the six individual SINS components were not significantly associated with the cumulative incidence of an event (table 4). The multivariate analysis included all six SINS components as well as total radiotherapy dose. Only the component location was significantly associated with the cumulative incidence of an adverse event with a hazard ratio of 0.54 (95%CI 0.30-0.96, p=0.04) (table 4).
Table 4. Results of the univariate and multivariate Fine and Gray models. HR: hazard ratio. CI: confidence interval.
Univariate Fine and Gray
HR
95% CI
p-value
Location
Pain
Bone lesion
Spinal alignment Vertebral body collapse Posterolateral elements Total SINS score
Total dose
Location
Pain
Bone lesion
Spinal alignment Vertebral body collapse Posterolateral elements Total dose
DISCUSSION
0.71 0.42-1.31 0.22 0.53 0.22-1.13 0.09 1.90 0.33-10.89 0.47 1.01 0.53-1.95 0.97 0.96 0.61-1.51 0.85 1.28 0.86-1.90 0.22 2.01 0.73-5.62 0.19 1.02 0.96-1.07 0.53
0.54 0.30-0.96 0.04 0.40 0.13-1.17 0.10 1.32 0.25-7.02 0.75 1.06 0.46-2.40 0.90 0.64 0.32-1.29 0.21 1.15 0.64-2.08 0.65 1.01 0.95-1.08 0.79
VII
Multivariate Fine and Gray
HR
95% CI
p-value
In this single center retrospective study it is demonstrated how the Spinal Instability Neoplastic Score performs on an independent dataset consisting of 110 patients undergoing external beam radiation therapy for symptomatic spinal bone metastases. Competing risk analysis showed that the SINS classification was not
105