Objective: To explore the outcome and prognostic factors of recursive partitioning analysis (RPA) Class III brain metastatic patients treated with stereotactic radiotherapy (SRT).
Materials and methods: Fifty-six consecutive patients with metastatic brain tumors and Karnofsky performance scale (KPS) scores <70 treated with SRT from January 2008 to October 2013 were involved in the analysis. Twenty-five patients (44.6%) were with symptomatic extracranial lesions (SELs), and the other 31 patients (55.4%) were without SELs. The detailed follow-up data of KPS scores were available in 44 patients. The KPS score drop time (KDT) was calculated as the time between SRT and 10 points drop of KPS scores compared to the baseline. Kaplan-Meier and Cox proportional hazards regression analyses were performed for univariate and multivariate analyses.
Results: The median overall survival time was 5.0 months (95% confidence interval [CI] 3.42-6.59) for the whole group. In multivariate analysis, the presence of SELs (P = 0.007, relative risk = 4.44, 95% CI 1.036-20.818) was the independent prognosis factor for survival. Median survival time was 3 months for the patients with SELs, 8 months for the patients without SELs. The median KDT of the 44 patients was 3.0 months (95% CI, 1.927-4.073 months). Again only the presence of SELs (P = 0.001, OR = 6.622, 95% CI, 2.108-20.801) was significantly related to KDT in multivariate analysis. The median KDT of the patients with SELs was 1.5 months, which was 5 months for the patients without SELs.
Conclusion: The presence of SELs was a negative prognosis factor for the survival of RPA Class III brain metastatic patients. If RPA Class III brain metastatic patients were without SELs, SRT may be a reasonable treatment option, but if they had SELs, SRT may not be a reasonable treatment due to the short overall survival time and KDT.