Borderline decisions in brain-metastatic breast cancer: efficacy of multimodality treatments in breast cancer patients with very limited prognosis suffering from brain metastases.

Diagnosis of brain metastases (BM) has traditionally marked a turning point in treatment goals and prognosis of patients with breast cancer (BC). However, new systemic treatment options have both prevented the development and improved the prognosis of BC patients with BM. Stereotactic radiotherapy techniques (SRT) have widely replaced whole brain radiotherapy (WBRT) in patients with limited BM, yielding impressive survival at modest toxicity rates. However, based on established prognostic scores, many patients will not profit from modern SRT treatment concepts and, therefore, remain candidates for WBRT or best supportive care (BSC). Here, treatment decisions may be challenging. This study aims to describe clinical outcomes and explore potential prognostic factors in a single-center cohort of BC patients treated with WBRT comprising patients across the prognosis continuum reflecting everyday work routine. We retrospectively analyzed 108 patients diagnosed with BM who received WBRT between 2008 and 2020. The cohort included patients with a broad range of prognostic factors (≥ 60 years old, Karnofsky Performance Status [KPS] ≤ 70, >3 BM and a Graded Prognostic Assessment [GPA] score < 1 [anticipated life expectancy of < 4 months]), who were either treated with palliative WBRT alone or in case of focal symptoms attributed to individual location of metastasis with surgery, SRT, or WBRT + focal dose escalation ("boost"). Survival rates were estimated using the Kaplan-Meier method, and factors potentially affecting outcome were statistically assessed by the log-rank test. Uni- and multivariable Cox regression was used in survival analysis to estimate hazard ratios and to evaluate the influence of various variables on survival time. For all statistical procedures SPSS software (v. 26) was used. Median survival after diagnosis of BM was 4 months (95% CI 3-5 months). The most relevant negative prognostic factor within the group of GPA-classifiers was KPS (≤ 70), followed by the number of BM (>3). Metastasectomy displayed significantly improved survival in univariable analysis (log rank p = 0.008; Cox-Regression p = 0.018), but did not elicit as predictive for OS in multivariable analysis (p = 0.47). Furthermore, completion of the prescribed RT series had the strongest impact on OS in both univariable and multivariable Cox regression analyses (p < 0.00001) while delivery of a radiation boost in addition to WBRT exerted no significant benefit (Cox regression p = 0.358). Despite the continuous development of effective treatment strategies for BM, BC patients who are ineligible for innovative treatment modalities show poor survival rates. Following prognostic assessment may help to make treatment decisions. KPS and number of BM may be most important in this context. In our retrospective cohort surgical removal of BM and completion of RT series were associated with improved prognosis, while focal radiation dose escalation in addition to WBRT was not.