Abstract
Background
Neoadjuvant stereotactic radiosurgery (SRS) has emerged as a novel therapeutic option in managing patients with brain metastasis (BM). Neoadjuvant SRS can address the current limitations of postoperative SRS through accurate lesion delineation and improved radiation dosimetry. This study evaluated the role of neoadjuvant SRS in patients with BM.
Methods
A comprehensive literature search was conducted. Studies that assessed the local control (LC), distant brain failure (DBF), leptomeningeal disease (LMD), overall survival (OS), and adverse radiation effects (ARE) rates in neoadjuvant SRS in BM were included.
Results
Ten studies involving 791 patients were included. The meta-analysis revealed a pooled 12- and 24-month LC rate of 91% (95% CI: 84%–96%) and 89% (95% CI: 81%–94%), respectively. The meta-analysis revealed 12- and 24-month DBF rates of 35% (95% CI: 31%–39%) and 44% (95% CI: 40%–47%), respectively. The meta-analysis exhibited a pooled 12- and 24-month LMD rate of 4% (95% CI: 2%–5%) and 5% (95% CI: 3%–7%), respectively. The meta-analysis showed a pooled 12- and 24-month ARE rate of 4% (95% CI: 3%–6%) and 5% (95% CI: 3%–7%), respectively. The meta-regression demonstrated that hypofractionated SRS (HF-SRS) was associated with a more favorable LC rate than single-fraction SRS (SF-SRS).
Conclusions
Neoadjuvant SRS is a promising option for BMs, offering favorable outcomes and a low toxicity profile. HF-SRS may be linked to a more favorable outcome than SF-SRS. Prospective randomized trials are necessary to compare neoadjuvant and postoperative SRS alongside evaluations of SF-SRS and HF-SRS.
Keywords: brain metastasis, leptomeningeal disease, neoadjuvant, preoperative, radiosurgery
Key Points.
Neoadjuvant SRS achieved high 12- and 24-month local control (LC) rates of 91% and 89%, respectively, with low leptomeningeal disease (8%) and adverse radiation effects (5%).
Hypofractionated SRS was associated with significantly better LC than single-fraction SRS, as determined by meta-regression analysis.
Compared to postoperative SRS, neoadjuvant SRS may allow for better lesion targeting, reduced leptomeningeal spread, and earlier systemic therapy, supporting its use in clinical practice.
Importance of the Study.
This study provides the latest meta-analysis assessing neoadjuvant SRS for BMs, including results from 838 patients across 10 studies. Compared to previous literature, this study quantifies LC, DBF, LMD, OS, and toxicity outcomes with increased precision. Notably, it underscores that neoadjuvant SRS achieves high LC and low LMD rates, alongside a favorable safety profile. The study further identifies hypofractionated SRS as potentially superior to single-fraction SRS. These results highlight the clinical utility of neoadjuvant SRS and suggest that it may address key limitations of postoperative approaches. The outcomes have direct implications for optimizing treatment strategies and refining eligibility criteria, underscoring the need for prospective randomized trials to validate and expand upon these observations.
Brain metastases (BMs) are the most frequent neoplastic lesion of the brain that can develop in up to 30% of individuals with cancer during the disease course.1–3 The incidence of BMs has substantially increased during the past decade, mainly due to more prolonged survival of cancer patients with systemic treatment and advancements in imaging studies.3 Resection remains an essential option in the management of BM cases, especially in patients with large or symptomatic lesions.1,4 Traditionally, postoperative stereotactic radiosurgery (SRS) has been used to target the resection cavity, thereby mitigating the likelihood of recurrence.5–8 Nevertheless, postoperative SRS has been reported to be associated with an increased probability of leptomeningeal disease (LMD), an increased incidence of adverse radiation effects (ARE), and variability in lesion segmentation.9–11
To mitigate the drawbacks of postoperative SRS, neoadjuvant SRS has been proposed as an alternative therapeutic option. By irradiating the intact lesion before resection, neoadjuvant SRS may provide enhanced radiation dosimetry, reduced LMD rates, and improved local control (LC).12–21 The primary advantage of the neoadjuvant SRS lies in the accurate lesion segmentation prior to resection, which eliminates the challenges associated with delineating the postresection cavity.22 Several studies have extensively investigated the clinical and radiological outcomes of the neoadjuvant SRS in patients with BM.12–21 However, the efficacy and safety of neoadjuvant SRS remain controversial regarding the heterogeneity in design, individual selection, and therapeutic parameters.
This systematic review and meta-analysis comprehensively assesses the literature on neoadjuvant SRS in BM. The objective is to synthesize the current data on LC, distant brain failure (DBF), LMD, overall survival (OS), and ARE in neoadjuvant SRS followed by resection in BM patients. This meta-analysis provides an updated evaluation of the radiological and clinical outcomes of the neoadjuvant SRS for future research directions and therapeutic protocols in the context of BMs.
Methods
Objective
The evaluation of the role of neoadjuvant SRS was the objective of the current study that followed the “Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA)” guidelines.23
Search Strategy
On May 4, 2025, PubMed, Embase, Scopus, and Web of Science were searched using “Radiosurgery,” “Neoadjuvant,” and “Brain Metastasis” and their equivalents. Supplementary Table S1 provides the search queries for each database.
Eligibility Criteria
The PICO framework is provided in Supplementary Table S2. Inclusion criteria:
The application of SRS modalities, including Gamma Knife radiosurgery (GKRS), CyberKnife radiosurgery (CKRS), and linear accelerator (LINAC), typically involves up to 5 sessions.
Report on LC, DBF, LMD, OS, and ARE for neoadjuvant SRS followed by resection in BM patients.
Studies involving at least 5 patients with neoadjuvant SRS followed by resection.
Exclusion criteria:
Reviews, book chapters, conference abstracts, preprints, and editorials.
Inability to separate the data of BM from other lesions
Lack of report on LC, DBF, LMD, OS, or ARE for neoadjuvant SRS patients
Overlapping participants
Studies that were published in a language other than English
Study Selection Process, Data Extraction, and Risk of Bias Assessment
Two authors (B.M. and M.H.) conducted the screening using Covidence Software, while a third author (I.M.) addressed the conflicts. Duplicates were managed by Covidence. Two authors (B.H. and M.H.) conducted the data extraction, while S.T. addressed the conflicts. The extracted variables are provided in Supplementary Table S3. The risk of bias (RoB) assessment was conducted by B.H., I.M., and M.H. using the Risk of Bias in Nonrandomized Studies—of Interventions tool (ROBINS-I).24
Statistical Analysis
Statistical analyses were conducted in R (v. 4.4.2) using the meta, metafor, and ggplot packages. Pooled proportions with 95% CIs were calculated through inverse-variance weighted random-effects meta-analysis, employing the Hartung-Knapp adjustment and Freeman-Tukey double arcsine transformation (PFT method). Where applicable, generalized linear mixed models were utilized with logit transformations (PLOGIT) for sensitivity analyses. The primary endpoints included pooled estimates for LC, 12-month LC, 24-month LC, DBF, 12-month DBF, 24-month DBF, LMD, 12-month LMD, 24-month LMD, 12-month OS, 24-month OS, ARE, 12-month ARE, and 24-month ARE. Heterogeneity was evaluated using the τ² statistic, and leave-one-out sensitivity analyses were performed to assess the robustness of the results. Publication bias was evaluated through funnel plot symmetry and Egger’s test. Meta-regression was employed to identify potential sources of heterogeneity. If a variable was reported as a median, it was converted to a mean using the method of Luo et al.25 The GRADE approach was used to assess the certainty of evidence, considering the risk of bias, inconsistency, indirectness, imprecision, and publication bias, as outlined in the GRADE guidelines.26
Results
Study Selection Process
The search identified 1698 studies, of which 592 were determined to be duplicates and were excluded before the initial screening (Figure 1). Consequently, 1106 studies were included in the screening process. During the title and abstract screen step, 962 were irrelevant and excluded. Thus, 140 studies were included in the full-text screening. During the full-text screening, 140 were excluded from these 3 due to stimulated neoadjuvant SRS,27–29 and 15 due to overlapping participants.3,10,30–42 Ultimately, 10 studies were included.12–21
Figure 1.
PRISMA flowchart.
Baseline Characteristics
Ten studies involving 791 patients who received neoadjuvant SRS were included (Table 1). The study designs were prospective in 50% (5/10), prospective plus retrospective in 20% (2/10), and retrospective in 30% (3/10) of the studies. The mean age ranged from 54.3 to 66.4 years. Males accounted for 45.9% (363/791) of the cohort, while females comprised 54.1% (428/791). Lung cancer (44.8%, 353/788) was the most common source of the brain disease, followed by breast cancer (14.2%, 112/788) and melanoma (13%, 102/788). Extracranial control was observed in 50.6% (316/624) of the cases. The LINAC (74.8%, 532/711) was the most common SRS modality, followed by GKRS (22.9%, 163/711) and CKRS (2.3%, 16/711) (Table 2). The mean prescribed dose ranged from 12 to 21 Gy in the SF-SRS and 24 to 30.7 Gy in 2-5 sessions in the HF-SRS. Gross total resection (GTR) was achieved in 95.7% (695/726) of the cases.
Table 1.
Baseline Characteristics of Patients
| Study | Year | Country | Recruitment period | Design | Patients | Mean age | Gender (M/F) | Lung | Melanoma | Breast | GI | RCC | Other brain mets | Extracranial controlled | Concurrent systemic therapy | Chemotherapy | Immunotherapy | Targeted therapy | Prior WBRT |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Udovicich et al. | 2025 | Australia | 2012–2013 | P + R | 179 | 63.2 | 77/102 | 78 | 31 | 21 | 15 | 7 | 27 | 108 | 71 | 22 | 24 | 22 | 3 |
| Higazy et al. | 2025 | Canada/Germany | 2018–2022 | P | 35 | 58.9 | 19/16 | 14 | 0 | 3 | 7 | 2 | 9 | NA | NA | 8 | 3 | 2 | 0 |
| Jansen et al. | 2024 | USA | 2021–2023 | P | 21 | NA | 9/12 | 5 | 7 | 4 | 2 | 0 | 3 | NA | NA | NA | 5 | NA | NA |
| Mallela et al. | 2024 | USA | 2016–2020 | P | 26 | 64.5 | 13/13 | 9 | 4 | 4 | 4 | 2 | 3 | NA | NA | NA | 5 | NA | 2 |
| Murphy et al. | 2024 | USA | 2013–2022 | P | 35 | 60.3 | 12/23 | 20 | 1 | 5 | 6 | 0 | 3 | NA | NA | NA | 11 | 10 | 0 |
| Kutuk et al. | 2024 | USA | 2017–2022 | R | 39 | 62.2 | 17/22 | 17 | 2 | 8 | 1 | 0 | 8 | 32 | 32 | 23 | 20 | 12 | NA |
| Rogers et al. | 2024 | Switzerland | 2020–2022 | P | 13 | 62.2 | 4/9 | 5 | 2 | 1 | 5 | 0 | 0 | NA | 4 | 1 | 2 | 1 | 0 |
| Prabhu et al. | 2023 | USA | 2005–2021 | P + R | 404 | 60.6 | 190/214 | 192 | 50 | 62 | 28 | 29 | 43 | 168 | NA | NA | NA | NA | 0 |
| Deguchi et al. | 2022 | Japan | 2019–2021 | R | 20 | 66.4 | 14/6 | 9 | 1 | 0 | 5 | 2 | 3 | NA | NA | NA | NA | NA | NA |
| Vetlova et al. | 2017 | Russia | 2015–2017 | R | 19 | 54.3 | 8/11 | 4 | 4 | 4 | 2 | 4 | 1 | 8 | NA | NA | NA | NA | 0 |
Abbreviations: NA, not available; M, male; F, female; GI, gastrointestinal; RCC, renal cell carcinoma; WBRT, whole brain radiotherapy.
Table 2.
Characteristics of Stereotactic Radiosurgery
| Study | Tumor volume (cc) | GKRS | CKRS | LINAC | Prescribed dose (Gy) | No. fractions | Post-SRS GTR | Post-SRS STR | En bloc resection | Piecemeal resection | Time SRS to surgery (days) |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Udovicich et al. | 10.81 | 64 | 0 | 125 | S-SRS: 18 HF-SRS: 24-27.5 |
2.00 | 135 | 5 | 74 | 88 | 3.43 |
| Higazy et al. | 16.6 | 26 | 9 | 0 | 16.2 | 1.00 | 32 | 3 | 15 | 15 | 3.9 |
| Jansen et al. | 23.7 | NA | NA | NA | NA | 1.60 | NA | NA | NA | NA | 4.05 |
| Mallela et al. | 13.27 | NA | NA | NA | 17.3 | 1.00 | NA | NA | NA | NA | NA |
| Murphy et al. | NA | NA | NA | NA | 46012 | 1.00 | 36 | 0 | 0 | 14 | 2.5 |
| Kutuk et al. | NA | 39 | 0 | 0 | NA | 1.00 | 47 | 0 | NA | NA | 45 695 |
| Rogers et al. | 14.5 | 0 | 0 | 12 | 16 | 1.00 | 12 | 1 | NA | NA | 2.4 |
| Prabhu et al. | 10 | 34 | 7 | 375 | S-SRS: 15 HF-SRS: 24 |
1.46 | 397 | 19 | 99 | 229 | 2 |
| Deguchi et al. | 20.6 | 0 | 0 | 20 | 30.7 | 5.00 | 17 | 3 | 0 | 20 | 4 |
| Vetlova et al. | 20 | NA | NA | NA | 18.14 | 1.00 | 19 | 0 | NA | NA | 1 |
Abbreviations: NA, not available; GKRS, gamma knife radiosurgery; CKRS, cyberknife radiosurgery; LINAC, linear accelerator; GTR, gross total resection; STR, subtotal resection; SRS, stereotactic radiosurgery; SF-SRS, single-fraction stereotactic radiosurgery; HF-SRS, hypofractionated stereotactic radiosurgery.
Clinical and Radiological Outcomes
The follow-up duration ranged from 6.1 to 64 months (Table 3). The LC rate across the included studies varied from 82.9% to 100%. The 12- and 24-month LC rates ranged from 76.9% to 100% and 69.2% to 100%, respectively. The overall LMD, along with the 12-month and 24-month LMD rates, ranged from 0% to 9.5%, 0% to 11.4%, and 0% to 14.3% in the included cohorts. The DBF rate varied from 15.8% to 48.6%, while the 12- and 24-month DBF rates ranged from 30% to 46.2% and 30% to 61.5%, respectively. The 12- and 24-month OS rates ranged from 41.7% to 89.5% and 25.7% to 89.5%, respectively. The ARE rate varied from 0% to 11.1%, and the 12- and 24-month ARE rates ranged from 0% to 8.6% and 0% to 8.6%, respectively.
Table 3.
Clinical and Radiological Outcomes Following Preoperative SRS
| Study | Mean FU (months) | LC | 1-year LC | 2-yearLC | LMD | 1-year LMD | 2-year LMD | DF | 1-year DF | 2-year DF | 1-year OS | 2-year OS | ARE | 1-year ARE | 2-year ARE |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Udovicich et al. | 28.40 | 93.7% | 95.2% | 93.7% | 9.5% | 7.3% | 11.2% | 29.1% | 37.3% | 47.8% | 66.5% | 48% | 5.3% | 3.7% | 4.2% |
| Higazy et al. | 11.80 | 82.9% | 82.9% | 82.9% | 2.9% | 2.9% | 2.9% | 45.7% | 34.3% | 45.7% | 51.4% | 25.7% | 8.6% | 8.6% | 8.6% |
| Jansen et al. | 10.80 | 100% | 100% | 100% | 4.8% | 4.8% | 4.8% | NA | 38.1% | 47.6% | 66.7% | NA | 4.8% | 4.8% | 4.8% |
| Mallela et al. | 12.00 | NA | 76.9% | 69.2% | 3.8% | 3.8% | 3.8% | NA | NA | NA | 57.7% | 38.5% | 3.8% | 3.8% | NA |
| Murphy et al. | 64.00 | 83.3% | 77.8% | 77.8% | 2.8% | 0% | 0% | 48.6% | 45.7% | 48.6% | 58.3% | 44.4% | 11.1% | 8.3% | 8.3% |
| Kutuk et al. | 14.00 | NA | 92.3% | 87.2% | NA | 11.4% | 14.3% | NA | 46.2% | 61.5% | 82.1% | 64.1% | NA | 5.1% | 5.1% |
| Rogers et al. | 6.10 | 83.3% | 83.3% | 83.3% | 0% | 0% | 0% | 33.3% | NA | NA | 41.7% | NA | 8.3% | 8.3% | 8.3% |
| Prabhu et al. | 18.10 | NA | 88% | 86.3% | NA | 4.5% | 5.7% | NA | 32.9% | 41.1% | 59.9% | 41.6% | NA | 5.9% | 7.4% |
| Deguchi et al. | 9.10 | 95% | 95% | 95% | 5% | 5% | 5% | 30% | 30% | 30% | 50% | 50% | 0% | 0% | 0% |
| Vetlova et al. | 8.06 | 89.5% | 94.7% | 89.5% | 5.3% | 5.3% | 5.3% | 15.8% | NA | NA | 89.5% | 89.5% | 5.3% | 5.3% | 5.3% |
Abbreviations: NA, not available; FU, follow-up; LC, local control; LMD, leptomeningeal disease; DF, distant failure; OS, overall survival; ARE, adverse radiation effect; M, months.
Meta-Analysis of LC and DBF
Seven studies were included in the meta-analysis of LC rate (Figure 2A). The meta-analysis showed a pooled LC rate of 92% (95% CI: 84%–98%). The meta-regression indicated that a higher number of fractions (Estimate = 0.482, P = .049) and a higher percentage of melanoma cases (Estimate = 0.055, P = .017), prior whole brain radiotherapy (WBRT) (Estimate = 0.608, P = .013) were linked to higher LC rates (Supplementary Table S4). Ten studies were included in the meta-analysis of the 12-month LC rate (Figure 2B). The meta-analysis found a pooled 12-month LC rate of 91% (95% CI: 84%–96%). The meta-regression showed that a greater mean number of fractions (Estimate = 0.514, P = .048) was associated with higher 12-month LC rates (Supplementary Table S4). Ten studies were included in the meta-analysis of the 24-month LC rate (Figure 2C). The pooled 24-month LC rate was 89% (95% CI: 81%–94%). The meta-regression revealed that a higher mean number of fractions (Estimate = 0.616, P = 0.009) correlated with higher 24-month LC rates (Supplementary Table S4).
Figure 2.
Forest plots showing pooled proportions following neoadjuvant SRS for BMs: (A) overall LC, (B) 12-month LC, (C) 24-month LC, (D) overall DBF, (E) 12-month DBF, and (F) 24-month DBF.
Six studies were included in the meta-analysis of the DBF rate (Figure 2D). The meta-analysis revealed a pooled DBF rate of 34% (95% CI: 21%–47%). The meta-regression analysis indicated that melanoma patients (Estimate = -0.048, P = .007) were associated with lower DBF rates (Supplementary Table S4). Seven studies were also included in the meta-analysis of the 12-month DBF rate (Figure 2E). This meta-analysis indicated a 12-month DBF rate of 35% (95% CI: 31%–39%). Additionally, 7 studies formed the basis for the meta-analysis of the 24-month DBF rate (Figure 2F), which demonstrated a pooled 24-month DBF rate of 44% (95% CI: 40%–47%). The meta-regression did not identify any sources of heterogeneity for the 12- and 24-month DBF rates (Supplementary Table S4).
Meta-Analysis of Leptomeningeal Disease
Eight studies were included in the meta-analysis of LMD rate (Figure 3A). The meta-analysis showed a pooled LMD rate of 8% (95% CI: 6%–12%). The meta-regression did not show any sources of heterogeneity for LMD (Supplementary Table S4). Ten studies were included in the meta-analysis of the 12-month LMD rate (Figure 3B). The meta-analysis exhibited a pooled 12-month LMD rate of 4% (95% CI: 2%–5%). The meta-regression did not show any sources of heterogeneity for 12-month LMD (Supplementary Table S4). Ten studies were included in the meta-analysis of the 24-month LMD rate (Figure 3C). The meta-analysis showed a pooled 24-month LMD rate of 5% (95% CI: 3%–7%). The meta-regression analysis demonstrated that patients undergoing en bloc resection (Estimate = 0.032, P = .031) had higher 24-month LMD rates (Supplementary Table S4).
Figure 3.
Forest plots showing pooled proportions of LMD following neoadjuvant SRS for BMs: (A) overall LMD rate, (B) 12-month LMD rate, (C) 24-month LMD rate, (D) overall OS rate, (E) 12-month OS rate, (F) overall ARE rate, (G) 12-month ARE rate, and (H) 24-month ARE rate.
Meta-Analysis of Overall Survival
Ten studies were included in the meta-analysis of 12-month OS rates (Figure 3D). The meta-analysis showed a pooled 12-month OS rate of 64% (95% CI: 54%–73%). The meta-regression demonstrated that a higher mean number of days between SRS and resection (Estimate = 0.00, P = .016), a higher percentage of breast cancer patients (Estimate = 0.065, P = .008), a lower rate of gastrointestinal cancer (Estimate = -0.039, P = .011), and a higher percentage of GTR patients (Estimate = 0.083, P = .008) were associated with higher 12-month OS rates (Supplementary Table S4). Eight studies were included in the meta-analysis of 24-month OS rates (Figure 3E). The meta-analysis demonstrated a pooled 24-month OS rate of 50% (95% CI: 34%–66%). Meta-regression did not identify any possible source of heterogeneity for 24-month OS (Supplementary Table S4).
Meta-Analysis of ARE
Eight studies were included in the meta-analysis of ARE rate (Figure 3F). The meta-analysis showed a pooled ARE rate of 5% (95% CI: 2%–8%). Ten studies were included in the meta-analysis of the 12-month ARE rate (Figure 3G). The meta-analysis showed a pooled 12-month ARE rate of 4% (95% CI: 3%–6%). Nine studies were included in the meta-analysis of the 24-month ARE rate (Figure 3H). The meta-analysis showed a pooled 24-month ARE rate of 5% (95% CI: 4%–7%). The meta-regression did not identify any possible source of heterogeneity for ARE, 12-month ARE, and 24-month ARE rates meta-analyses (Supplementary Table S4).
Sensitivity Analysis
The sensitivity analysis for the LC rate showed that the results were robust (Supplementary Figure S1A). The sensitivity analysis for 12- and 24-month LC rates showed that the results were robust (Supplementary Figure S1B and C). The sensitivity analysis for DBF, 12-month DBF, and 24-month DBF rates revealed that the results were robust, with the pooled estimates remaining consistent (Supplementary Figure S1D–F). The sensitivity analyses of LMD, 12-month LMD, and 24-month LMD were also indicative of the robustness of the results (Supplementary Figure S1G–I). The sensitivity analyses of 12-month OS, 24-month OS, ARE, 12-month ARE, and 24-month ARE showed the robustness of the results, and none of the included studies had a significant impact on the pooled estimate (Supplementary Figure S1J–N).
Publication Bias
The funnel plots for LC, 24-month LC, DBF, and 24-month DBF had a symmetrical pattern, while 12-month LC and DBF showed a minor asymmetry (Supplementary Figure S2A–F). The corresponding Egger’s test was higher than 0.5 for all of them, showing no significant publication bias (Supplementary Table S5). Similarly, the funnel plots for the LMD, 12-month LMD, 24-month LMD, 12-month OS, 24-month OS, ARE, 12-month ARE, and 24-month ARE rates showed symmetrical patterns, and the Egger’s test showed no significant publication bias (Supplementary Figure S2G–N and Supplementary Table S5).
Risk of Bias Assessment
The RoB assessment of the included studies demonstrated that 9 were associated with moderate RoB and 2 were related to serious RoB (Supplementary Figure S3). Bias due to confounding was the primary source of heterogeneity, as some of the included studies were based on retrospective cohorts. The serious RoB was due to bias in outcome measurement in one study and bias due to confounding in the other. In general, the results of the current study should be interpreted carefully.
GRADE Assessment
The certainty level of evidence of 12- and 24-month LC, DBF, LMD, OS, and ARE was rated as moderate due to RoB, low heterogeneity, and no serious concerns regarding publication bias or indirectness (Supplementary Table S6). On the other hand, the LC, OS, and DBF were rated as low due to high heterogeneity and wide 95% CI (Supplementary Table S6).
Discussion
Neoadjuvant SRS has emerged as a promising therapeutic option in the management of individuals with BMs to mitigate the drawbacks of the postoperative SRS. This systematic review and meta-analysis highlight that the neoadjuvant SRS is associated with favorable clinical and radiological outcomes concurrent with low adverse events (Table 4). Neoadjuvant SRS achieved a pooled 24-month LC rate of 89% (95% CI: 81%–94%), 24-month DBF rate of 44% (95% CI: 40%–47%), 24-month LMD rate of 5% (95% CI: 3%–7%), and 24-month ARE rate of 5% (95% CI: 3%–7%). These findings suggest that neoadjuvant SRS can yield comparable outcomes and even superior results in some cases to postoperative SRS, indicating that this approach may be a promising treatment option for BM patients.
Table 4.
Summary of Meta-Analysis Results
| Meta-analysis | No. studies | Pooled estimate |
|---|---|---|
| LC | 7 | 92% (95% CI: 84%–98%) |
| 12-month LC | 10 | 91% (95% CI: 84%–96%) |
| 24-month LC | 10 | 89% (95% CI: 81%–94%) |
| DBF | 6 | 34% (95% CI: 21%–47%) |
| 12-month DBF | 6 | 35% (95% CI: 31%–39%) |
| 24-month DBF | 6 | 44% (95% CI: 40%–47%) |
| LMD | 8 | 8% (95% CI: 6%–12%) |
| 12-month LMD | 10 | 4% (95% CI: 2%–5%) |
| 24-month LMD | 10 | 5% (95% CI: 3%–7%) |
| 12-month OS | 10 | 64% (95% CI: 55%–73%) |
| 24-month OS | 8 | 50% (95% CI: 34%–66%) |
| ARE | 8 | 5% (95% CI: 2%–8%) |
| 12-month ARE | 10 | 4% (95% CI: 3%–6%) |
| 24-month ARE | 9 | 5% (95% CI: 4%–7%) |
Abbreviations: LC, local control; LMD, leptomeningeal disease; DF, distant failure; OS, overall survival; ARE, adverse radiation effect.
The paradigm shift from postoperative to neoadjuvant SRS is driven by accumulating evidence underscoring the biological and logistical advantages of irradiating the intact lesion. Postoperative SRS is associated with several limitations, including challenges in precise target delineation due to postresection alterations, delayed initiation of adjuvant therapy, and a potentially increased likelihood of leptomeningeal dissemination resulting from intervention manipulation.22,43 On the other hand, neoadjuvant SRS is associated with an enhanced radiation dosimetry through accurate lesion delineation.12–21,42 Perlow et al. demonstrated that the LC rates were 96% in postoperative fractionated radiotherapy (FSRT), 100% in neoadjuvant FSRT, 100% in postoperative FSRT, and 96% in neoadjuvant SRS groups (P = .5).32 They also indicated that the LMD rate was 5.3% in postoperative FSRT, 1.9% in neoadjuvant FSRT, 0% in postoperative SRS, and 3.7% in neoadjuvant SRS groups (P = .9).32 Patel et al. demonstrated that there was no significant difference between neoadjuvant and postoperative SRS regarding OS and local recurrence; however, 2-year ARE (Neoadjuvant = 3.2% vs. Postoperative = 16.6%, P = .01) and 2-year ARE (Neoadjuvant = 4.9% vs. Postoperative = 16.4%, P = .01) were significantly higher in the postoperative SRS groups.41
A crucial aspect of applying neoadjuvant SRS is deciding between single-fraction (SF-SRS) and hypofractionated (HF-SRS) approaches. While the SF-SRS approach is more convenient, it can be correlated with suboptimal biological efficacy in larger lesions and an increased probability of ARE development.30 Prabhu et al. in the PROPS-BM cohort demonstrated that SF-SRS is associated with higher 2-year local recurrence rates (SF-SRS = 16.3% vs. HF-SRS = 2.9%, P = .004) without any differences in 24-month ARE (SF-SRS = 6.7% vs. HF-SRS = 10.7%, P = .16), LMD (SF-SRS = 6.1% vs. HF-SRS = 3.8%, P = .94), or OS (SF-SRS = 40.4% vs. HF-SRS = 48.7%, P = .44) rates.30 Our meta-regression also underscored that 12- and 24-month LC rates were higher in the HF-SRS than SF-SRS (Supplementary Table S4). The superiority of HF-SRS in LC, as observed in our results, may arise from its ability to deliver a higher biologically effective dose compared to SF-SRS, especially in larger lesions where single-dose delivery is limited by the risk of radiation toxicity to surrounding tissues. Fractionation enables dose escalation while reducing the risk of AREs, potentially enhancing lesion control. However, this benefit must be balanced against the challenges of administering multiple SRS sessions. Many patients receiving neoadjuvant SRS are acutely symptomatic or admitted with neurological compromise, and HF-SRS might delay definitive surgery, creating a clinical dilemma. Therefore, choosing between SF-SRS and HF-SRS should be tailored to each case, considering tumor size, location, the urgency of surgery, and available institutional resources.
The integration of the neoadjuvant SRS in the clinical workflow should be meticulous. Individuals should be assessed by radiation oncologists and neurosurgeons regarding the feasibility of neoadjuvant SRS. In addition, the optimal timing between SRS and resection should be discussed as our meta-regression demonstrated that a longer interval is associated with high LC rates, potentially because of improved radiation-induced lesion response prior to surgical resection. In addition, despite the less convenient nature of HF-SRS, this approach is associated with more favorable clinical and radiological outcomes. To effectively implement neoadjuvant SRS, centers must develop standardized therapeutic and follow-up protocols to optimize treatment strategy and improve patient outcomes.
Future studies should focus on comparing neoadjuvant and postoperative SRS in large, prospective, randomized clinical trials, concurrent with comparing HF-SRS and SF-SRS to determine the indications for each approach. In addition, studies should provide the data separately for the underlying BM histologies, as our data suggested that melanoma cancer, which is considered radioresistant, is associated with better results with neoadjuvant SRS. Also, the en bloc and non-en bloc approaches following SRS should be compared to determine the indication for each technique. The synergistic or antagonistic effects of neoadjuvant SRS with systemic treatments should be evaluated in future studies.
Our study has several limitations. The presence of a retrospective nature in some of the included studies is one of the limitations that may introduce patient selection bias and restrict the generalizability of the results. Heterogeneity is another limitation of our study, stemming from variations in lesion origin, SRS characteristics (such as the number of fractions), and follow-up protocols. We conducted meta-regression to overcome this limitation. Variations in the follow-up durations are another limitation that may impact the comparison of LC, LMD, and ARE. Another limitation is the variation in how studies report ARE. While some specify symptomatic AREs, others do not distinguish between symptomatic and asymptomatic findings. Therefore, we combined all reported AREs regardless of symptom status, which may lead to inconsistencies in interpreting toxicity outcomes.
Conclusion
The findings of this study suggest that neoadjuvant SRS is an efficient and safe approach in managing BM patients. We demonstrated that neoadjuvant SRS is associated with improved LC, DBF, LMD, ARE, and OS rates. Neoadjuvant SRS provides comparable, and in some cases even more favorable, outcomes than postoperative SRS; however, more double-arm studies are needed to determine the indications for each approach. HF-SRS may yield more favorable outcomes than the SF-SRS approach, and future studies should also compare these 2 SRS techniques. These findings underscore the necessity of prospective randomized clinical trials to validate our results and identify optimal candidates for neoadjuvant SF-SRS, neoadjuvant HF-SRS, and postinterventional SRS in the context of BM.
Supplementary Material
Contributor Information
Bardia Hajikarimloo, Department of Neurological Surgery, Shohada Tajrish Hospital, Shahid Beheshti University of Medical Sciences, Tehran, Iran.
Mohammad Amin Habibi, Department of Neurosurgery, Shariati Hospital, Tehran University of Medical Sciences, Tehran, Iran.
Ibrahim Mohammadzadeh, Skull Base Research Center, Loghman-Hakim Hospital, Shahid Beheshti University of Medical Sciences, Tehran, Iran.
Salem M Tos, Department of Neurological Surgery, University of Virginia, Charlottesville, Virginia, USA.
Ali Mortezaei, Department of Neurosurgery, Gonabad University of Medical Sciences, Gonabad, Iran.
Saleh Behzadi, Student Research Committee, Rafsanjan University of Medical Sciences, Rafsanjan, Iran.
Parsa Lorestani, Students Research Committee, Kermanshah University of Medical Sciences, Kermanshah, Iran.
Kiarash Soltani, Department of Neurological Surgery, Shohada Tajrish Hospital, Shahid Beheshti University of Medical Sciences, Tehran, Iran.
Mohammad Ali Nazari, Department of Neurosurgery, Iran University of Medical Sciences, Tehran, Iran.
Azin Ebrahimi, Department of Neurological Surgery, Shohada Tajrish Hospital, Shahid Beheshti University of Medical Sciences, Tehran, Iran.
Mohammadamin Sabbagh Alvani, Department of Neurological Surgery, Shohada Tajrish Hospital, Shahid Beheshti University of Medical Sciences, Tehran, Iran.
Conflict of interest statement. None declared.
Funding
None declared.
Author Contributions
Conceptualization: B.H. and M.H.; Methodology: B.H., M.H., and S.T.; Literature Search: B.H., M.H., and I.M.; Data Extraction: B.H., M.H., and I.M.; Risk of Bias Assessment: B.H., M.H., and I.M.; Statistical Analysis: B.H.; Writing—original draft: S.B., P.L., K.S., and M.N.; Writing—review & editing: B.H. and A.E., Supervision: B.H.; Project administration: B.H.
Data Availability
The data supporting this study’s findings are available from the corresponding author upon reasonable request.
References
- 1. Moravan MJ, Fecci PE, Anders CK, et al. Current multidisciplinary management of brain metastases. Cancer. 2020;126(7):1390–1406. [DOI] [PubMed] [Google Scholar]
- 2. Sacks P, Rahman M. Epidemiology of brain metastases. Neurosurg Clin N Am. 2020;31(4):481–488. [DOI] [PubMed] [Google Scholar]
- 3. Udovicich C, Ng SP, Tange D, Bailey N, Haghighi N. From postoperative to preoperative: a case series of hypofractionated and single-fraction neoadjuvant stereotactic radiosurgery for brain metastases. Oper Neurosurg. 2022;22(4):208–214. [DOI] [PubMed] [Google Scholar]
- 4. Mitchell DK, Kwon HJ, Kubica PA, et al. Brain metastases: an update on the multi-disciplinary approach of clinical management. Neurochirurgie. 2022;68(1):69–85. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Akanda ZZ, Hong W, Nahavandi S, et al. Post-operative stereotactic radiosurgery following excision of brain metastases: a systematic review and meta-analysis. Radiother Oncol. 2020;142:27–35. [DOI] [PubMed] [Google Scholar]
- 6. Marchan EM, Peterson J, Sio TT, et al. Postoperative cavity stereotactic radiosurgery for brain metastases. Front Oncol. 2018;8(1):342. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Dohm AE, Hughes R, Wheless W, et al. Surgical resection and postoperative radiosurgery versus staged radiosurgery for large brain metastases. J Neurooncol. 2018;140(3):749–756. [DOI] [PubMed] [Google Scholar]
- 8. Bander ED, El Ahmadieh TY, Chen J, et al. Outcomes following early postoperative adjuvant radiosurgery for brain metastases. JAMA Netw Open. 2023;6(10):e2340654. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Redmond KJ, De Salles AAF, Fariselli L, et al. Stereotactic radiosurgery for postoperative metastatic surgical cavities: a critical review and international stereotactic radiosurgery society (ISRS) practice guidelines. Int J Radiat Oncol Biol Phys. 2021;111(1):68–80. [DOI] [PubMed] [Google Scholar]
- 10. Patel KR, Burri SH, Boselli D, et al. Comparing pre-operative stereotactic radiosurgery (SRS) to post-operative whole brain radiation therapy (WBRT) for resectable brain metastases: a multi-institutional analysis. J Neurooncol. 2017;131(3):611–618. [DOI] [PubMed] [Google Scholar]
- 11. Soliman H, Ruschin M, Angelov L, et al. Consensus contouring guidelines for postoperative completely resected cavity stereotactic radiosurgery for brain metastases. Int J Radiat Oncol Biol Phys. 2018;100(2):436–442. [DOI] [PubMed] [Google Scholar]
- 12. Udovicich C, Koo K, Michael Bryant J, et al. ; INTERNEO Investigators. International collaboration of neoadjuvant stereotactic radiosurgery for brain metastases: The INTERNEO individual patient data pooled analysis. Radiother Oncol. 2025;202(110641):110641. [DOI] [PubMed] [Google Scholar]
- 13. Higazy R, Li D, Lau R, et al. Neoadjuvant stereotactic radiosurgery for large brain metastases: an international, multicenter, single-arm phase II trial. Neurosurg. 2025;1. [DOI] [PubMed] [Google Scholar]
- 14. Jansen CS, Pagadala MS, Cardenas MA, et al. Pre-operative stereotactic radiosurgery and peri-operative dexamethasone for resectable brain metastases: a two-arm pilot study evaluating clinical outcomes and immunological correlates. Nat Commun. 2024;15(1):8854. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Mallela AN, Iheagwara UK, Fogg D, et al. Preoperative stereotactic radiosurgery for cerebral metastases: safe, effective, and decreases steroid dependency. J Neurosurg. 2024;141(5):1332–1342. [DOI] [PubMed] [Google Scholar]
- 16. Murphy ES, Yang K, Suh JH, et al. Phase I trial of dose escalation for preoperative stereotactic radiosurgery for patients with large brain metastases. Neuro-Oncology. 2024;26(9):1651–1659. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Kutuk T, Zhang Y, Akdemir EY, et al. Comparative evaluation of outcomes amongst different radiosurgery management paradigms for patients with large brain metastasis. J Neurooncol. 2024;169(1):105–117. [DOI] [PubMed] [Google Scholar]
- 18. Rogers S, Schwyzer L, Lomax N, et al. Preoperative radiosurgery for brain metastases (PREOP-1): a feasibility trial. Clin Transl Radiat Oncol [Internet]. 2024;47(100798):100798. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Prabhu RS, Akinyelu T, Vaslow ZK, et al. Risk factors for progression and toxic effects after preoperative stereotactic radiosurgery for patients with resected Brain Metastases. JAMA Oncol [Internet]. 2023;9(8):1066–1073. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Deguchi S, Mitsuya K, Yasui K, et al. Neoadjuvant fractionated stereotactic radiotherapy followed by piecemeal resection of brain metastasis: a case series of 20 patients. Int J Clin Oncol. 2022;27(3):481–487. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Vetlova E, Golbin DA, Golanov AV, et al. Preoperative stereotactic radiosurgery of brain metastases: preliminary results. Cureus 2017;9(12):e1987. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Crompton D, Koffler D, Fekrmandi F, et al. Preoperative stereotactic radiosurgery as neoadjuvant therapy for resectable brain tumors. J Neurooncol. 2023;165(1):21–28. [DOI] [PubMed] [Google Scholar]
- 23. Page MJ, McKenzie JE, Bossuyt PM, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. PLoS Med. 2021;18(3):e1003583. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Sterne JA, Hernán MA, Reeves BC, et al. ROBINS-I: a tool for assessing risk of bias in non-randomised studies of interventions. BMJ (Clinical research ed.). 2016;355:i4919. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Luo D, Wan X, Liu J, Tong T. Optimally estimating the sample mean from the sample size, median, mid-range, and/or mid-quartile range. Stat Methods Med Res. 2018;27(6):1785–1805. [DOI] [PubMed] [Google Scholar]
- 26. Schünemann HJ, Brennan S, Akl EA, et al. The development methods of official GRADE articles and requirements for claiming the use of GRADE – a statement by the GRADE guidance group. J Clin Epidemiol. 2023;159(1):79–84. [DOI] [PubMed] [Google Scholar]
- 27. Cheok SK, Yu C, Feng JJ, et al. Comparison of preoperative versus postoperative treatment dosimetry plans of single-fraction stereotactic radiosurgery for surgically resected brain metastases. Neurosurg Focus. 2023;55(2):E9. [DOI] [PubMed] [Google Scholar]
- 28. El Shafie RA, Tonndorf-Martini E, Schmitt D, et al. Pre-operative versus post-operative radiosurgery of brain metastases-volumetric and dosimetric impact of treatment sequence and margin concept. Cancers. 2019;11(3):294. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Vellayappan BA, Doody J, Vandervoort E, et al. Pre-operative versus post-operative radiosurgery for brain metastasis: effects on treatment volume and inter-observer variability. J Radiosurg SBRT. 2018;5(2):89–97. [PMC free article] [PubMed] [Google Scholar]
- 30. Prabhu RS, Akinyelu T, Vaslow ZK, et al. Single-fraction versus fractionated preoperative radiosurgery for resected brain metastases: a PROPS-BM international multicenter cohort study. Int J Radiat Oncol Biol Phys. 2024;118(3):650–661. [DOI] [PubMed] [Google Scholar]
- 31. Prabhu RS, Akinyelu T, Vaslow ZK, et al. Standard dose vs. reduced dose single fraction preoperative radiosurgery for resected brain metastases: a PROPS-BM international multicenter cohort study. Int J Radiat Oncol Biol Phys. 2024;120(2):e266–e267. [DOI] [PubMed] [Google Scholar]
- 32. Perlow HK, Ho C, Matsui JK, et al. Comparing pre-operative versus post-operative single and multi-fraction stereotactic radiotherapy for patients with resectable brain metastases. Clinical and translational radiation oncology. 2023;38(1):117–122. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Li YD, Coxon AT, Huang J, et al. Neoadjuvant stereotactic radiosurgery for brain metastases: a new paradigm. Neurosurg Focus. 2022;53(5):E8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Palmer JD, Perlow HK, Matsui JK, et al. Fractionated pre-operative stereotactic radiotherapy for patients with brain metastases: a multi-institutional analysis. J Neurooncol. 2022;159(2):389–395. [DOI] [PubMed] [Google Scholar]
- 35. Kotecha R, Tonse R, Menendez MAR, et al. Evaluation of the impact of pre-operative stereotactic radiotherapy on the acute changes in histopathologic and immune marker profiles of brain metastases. Sci Rep. 2022;12(1):4567. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Prabhu RS, Dhakal R, Vaslow ZK, et al. Preoperative radiosurgery for resected brain metastases: the PROPS-BM multicenter cohort study. Int J Radiat Oncol Biol Phys. 2021;111(3):764–772. [DOI] [PubMed] [Google Scholar]
- 37. Takami H, Nassiri F, Moraes FY, et al. A phase II study of neoadjuvant stereotactic radiosurgery for large brain metastases: clinical trial protocol. Neurosurgery [Internet]. 2020;87(2):403–407. [DOI] [PubMed] [Google Scholar]
- 38. Prabhu RS, Miller KR, Asher AL, et al. Preoperative stereotactic radiosurgery before planned resection of brain metastases: updated analysis of efficacy and toxicity of a novel treatment paradigm. J Neurosurg. 2019;131(5):1387–1394. [DOI] [PubMed] [Google Scholar]
- 39. Patel AR, Nedzi L, Lau S, et al. Neoadjuvant stereotactic radiosurgery before surgical resection of cerebral metastases. World Neurosurg. 2018;120(1):e480–e487. [DOI] [PubMed] [Google Scholar]
- 40. Prabhu RS, Press RH, Patel KR, et al. Single-fraction stereotactic radiosurgery (SRS) alone versus surgical resection and SRS for large brain metastases: a multi-institutional analysis. Int J Radiat Oncol Biol Phys. 2017;99(2):459–467. [DOI] [PubMed] [Google Scholar]
- 41. Patel KR, Burri SH, Asher AL, et al. Comparing preoperative with postoperative stereotactic radiosurgery for resectable brain metastases: a multi-institutional analysis. Neurosurgery. 2016;79(2):279–285. [DOI] [PubMed] [Google Scholar]
- 42. Asher AL, Burri SH, Wiggins WF, et al. A new treatment paradigm: neoadjuvant radiosurgery before surgical resection of brain metastases with analysis of local tumor recurrence. Int J Radiat Oncol Biol Phys. 2014;88(4):899–906. [DOI] [PubMed] [Google Scholar]
- 43. Routman DM, Yan E, Vora S, et al. Preoperative stereotactic radiosurgery for brain metastases. Front Neurol. 2018;9(1):959. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data supporting this study’s findings are available from the corresponding author upon reasonable request.