Application of a multi-institutional nomogram predicting salvage whole brain radiation-free survival to patients treated with postoperative stereotactic radiotherapy for brain metastases

Purpose: The ultimate goal of stereotactic radiotherapy (SRT) of brain metastases (BM) is to avoid or postpone whole brain radiotherapy (WBRT). A nomogram based on multi-institutional data was developed by Gorovets, et al. to estimate the 6 and 12-months WBRT-free survival (WFS). The aim of the current retrospective study was to validate the nomogram in a cohort of postoperative BM patients treated with adjuvant SRT. Material and methods: We reviewed the data of 68 patients treated between 2008–2017 with postoperative SRT for BM. The primary endpoint was the WFS. The receiver operating characteristic curve and area under the curve (AUC) were calculated for both 6- and 12-months time points. Results: After a median follow-up of 64 months, the 1-year cumulative incidence of local and distant brain relapse rates were 15% [95% CI = 8–26%] and 34% [95% CI = 24–48%], respectively. At recurrence, repeated SRT or salvage WBRT were applied in 33% and 57% cases, respectively. The WFS rates at 6 and 12 months were 88% [95% CI = 81–97%] and 67% [95% CI = 56–81%], respectively. Using the Gorovets nomogram, the 6 months rates were overestimated while they were accurate at 12 months. AUC values were 0.47 and 0.62 for the 6- and 12-months respectively. Overall, Harrell's concordance index was 0.54. Conclusion: This nomogram-predicted well the 12 months WFS but its discriminative power was quite low. This underlines the limits of this kind of predictive tool and leads us to consider the use of big data analysis in the future. © 2020 Société française de radiothérapie oncologique (SFRO)