Approximately20% of patients with extracranial cancer develop brain metastases (BMs), representing the most common type of intracranial tumor.1These lesions usually reduce lifespan and quality of life while they increase focal neurological deficits and cognitive impairment due to mass effect, which is dependent on the size and location of the BM.2Furthermore, several BMs present as large BMs (LBMs), considered as a cutoff of ≥ 4 cm3in volume and ≥ 2 cm in maximum diameter.3Although most can be accessed surgically, complex morphology, eloquent location, or patient comorbidities can require alternative strategies to optimize local control, enhance the volumetric reduction, and decrease the adverse effects of the treatment.3Lately, hypofractionated stereotactic radiosurgery (SRS) has been increasingly implemented in these situations to control LBMs, with a low adverse effects rate.4–6
A similar approach called multisession or staged SRS is being incorporated for LBMs, with a smaller radiation area and lower radiation-related toxicity. It consists of treating the patient with radiation in separate stages, with a varying time interval and dose (in Gy) between treatments, allowing practitioners to modify to a smaller radiation field due to the shrinkage in tumoral tissue.7–12This approach has proven to be both an effective and safe treatment modality for LBMs, with excellent local control in the long term, especially regarding the most frequent staging strategies (2-stage and 3-stage SRS).13–16
According to American Society of Clinical Oncology–Society for Neuro-Oncology–American Society for Therapeutic Radiology and Oncology (ASCO-SNO-ASTRO) guidelines, SRS should be offered to patients with brain metastases < 3 or 4 cm in diameter, because it achieves equal tumoral response to whole-brain radiation therapy (WBRT) with decreased cognitive deterioration.17The available literature on staged SRS is mainly reported via retrospective observational studies, showing an uncertain quality of evidence. Given that, no methodological or expert consensus has been achieved on the use of staged SRS for LBMs. Therefore, we aimed to determine the effectiveness and safety of 2-stage SRS for LBMs, in order to optimize patients’ postoperative course.
开云体育世界杯赔率
Protocol and Registration
This study was performed following the recommendations of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA).18The protocol for this review was submitted to PROSPERO prior to initiation, with the registration number CRD42022384463 (PROSPERO).19
Literature Search and Study Selection
The systematic search was carried out in 6 databases: PubMed/MEDLINE, Scopus, Web of Science, Embase, Cochrane (OvidSP), and Google Scholar for works published up to December 14, 2022.
The search strategy was adapted for each database and included the Medical Subject Headings (MeSH) terms for "brain neoplasms/secondary," "neoplasms metastasis," and related words for "radiosurgery" and "proton therapy." The final search strategy is shown inSupplementary Table 1.
The inclusion criteria were as follows: 1) diagnosis of LBMs (confirmed by histopathology or imaging); 2) adult patients (≥ 18 years old); 3) treatment with time-staged SRS due to surgical incompatibility; and 4) randomized and nonrandomized clinical trials, cohorts, case-control studies, and case series. No language restriction was implemented. We excluded case reports, animal studies, systematic reviews, or other type of overviews (e.g., mapping review, umbrella review, scoping review), and abstracts.
Duplicates obtained in the initial search were removed using Endnote X9 software.20In due course, screening by title and abstract was performed to identify eligible articles. Potentially relevant articles were evaluated in full text to determine their final eligibility according to the inclusion criteria. The selection process was carried out by two authors (F.T. and K.Z.) in an independent fashion. In case of discrepancies, a third author (M.A.M.) achieved consensus. The list of excluded articles is shown inSupplementary Table 2.
Data Extraction and Quality Evaluation
Data extraction from each included article was performed by the same two authors who did the selection (F.T. and K.Z.), collecting the following data: author; year; country; study design; number of participants; baseline characteristics (demographic, clinical, and tumor); and treatment characteristics variables. The third author (M.A.M.) evaluated the data extracted before proceeding to the analysis.
The risk of bias and quality of the studies were assessed using two different risk of bias assessment tools. The Risk of Bias in Nonrandomized Studies–of Interventions (ROBINS-I)21was used to assess cohorts and nonrandomized clinical trials by classifying them in low, moderate, serious, and critical risk groups. Additionally, the Joanna Briggs Institute risk of bias assessment tool was used for case series.22
Outcomes and Statistical Analyses
The primary volumetric outcome was the mean total volume reduction, defined as the absolute difference in LBMs volume (in cm3) between the baseline and last follow-up measurement. Secondary volumetric outcomes included Gamma Knife radiosurgery (GKRS) first session (GKRS1) volume reduction and GKRS second session (GKRS2) volume reduction. For that purpose, we calculated and reported the mean and standard deviation of the volume at three time points (baseline, after GKRS1, and at last follow-up), as well as the relative (%) volume reduction. The primary effectiveness outcome was complete response of the LBMs, which—according to Response Assessment in Neuro-Oncology Brain Metastases (RANO-BM) criteria—is defined as the disappearance of all target lesions for a minimum of 4 weeks.23The primary safety outcome was neurological mortality, defined as death due to intracranial disease progression. Our secondary safety outcome was all-cause mortality.
我运行一个合成的数据分析n which a random-effects model was used to estimate pooled estimates quantitatively, according to the DerSimonian-Laird method.24A Freeman-Tukey double-arcsine transformation25was performed a priori to stabilize the proportion variances. The I2statistic was considered to express the degree of heterogeneity. Missing quantitative measures were manually calculated.26Resulting estimates were expressed using pooled rates and mean differences with their respective 95% confidence intervals. Only observational studies were included in the meta-analysis.
We conducted a subgroup analysis according to study design, type of recollection (i.e., retrospective or prospective), number of healthcare centers, predominant sex, predominant primary tumor, use of prior treatment, and predominant prior treatment. Neither a sensitivity analysis nor publication bias assessment were performed due to the low number of included studies in the quantitative synthesis.27All the statistical processing was performed using RStudio version 4.2.2.
Evidence Certainty Assessment
According to the Cochrane Handbook for Systematic Reviews of Interventions,28the certainty of the evidence of eligible studies in quantitative synthesis was assessed using the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) system.29The summary of findings table was built in the GRADEpro online tool.29
Results
Study Selection
A total of 1782 titles were identified during the initial search. Full-text assessment for eligibility criteria was evaluated for 30 studies. After its application, we excluded 16 studies (Supplementary Table 2) and included 14 studies in the final analysis (Fig. 1).
Selection process flowchart. The last row accounts for the fact that 2 studies involved a comparison between 2-stage and 3-stage GKRS. Data added to the PRISMA template (from Moher D, Liberati A, Tetzlaff J, Altman DG, The PRISMA Group. Preferred Reporting Items for Systematic Reviews and Meta-Analyses: The PRISMA Statement.PLoS Med.2009;6[7]:e1000097) under the terms of the Creative Commons Attribution (CC BY-NC 2.0) License (https://creativecommons.org/licenses/by/2.0).
Characteristics of the Studies
In the 14 studies, 958 patients received staged GKRS for a total of > 1019 LBMs (Table 1). Only Hasegawa et al.30and Serizawa et al.7made a direct comparison between the 2-stage and the 3-stage approach. The median age among patients at the time of diagnosis or first GKRS session ranged between 59 and 69.5 years old. Regarding the performance status, 9 studies7–9,11–13,30–32reported the Karnofsky Performance Scale score using a cutoff of < 80 to define disability for work activities, whereas 3 studies14,16,33reported the Eastern Cooperative Oncology Group (ECOG) Performance Status Scale. A cutoff of ≥ 2 was set to define disability for work activities. All the studies heterogeneously defined LBMs by assigning a cutoff for either the maximum diameter or the volume. Even so, 2 studies11,31did not set a clear definition of LBM. The most frequent primary tumor focus was lung cancer, specifically non–small cell lung cancer.7,13,14For the treatment prior to SRS, the predominant method was systemic therapy in 3 studies,8,13,14WBRT in 4 studies,11,12,16,33and resection surgery in 3 studies.9,16,32Uniquely, the use of an Ommaya reservoir was reported as primary treatment in 1 study.30
Baseline characteristics reported by the 14 included studies
| Authors & Year | Study Design | Sample Size | Male Pts, No. (%) | Median Age in Yrs (range) | No. Disabled (%) | Prior Primary Tx, No. (%) | Diam (cm) or Vol (cm3) Cutoff for LBM | No. LBMs | Primary Tumor Focus, No. (%) | Supratentorial Location, No. (%) | No. GKRS Sessions |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Cui et al., 202233 | Retro, SC, CA | 12 | 2 (16.67) | 59.5 (29–80)* | 5 (41.67)† | WBRT 1 (8.33) | ≥2 cm or >4 cm3 | 24 | Breast 6 (25) | NR | 2 |
| Damron et al., 202214 | Retro, SC, SA | 24 | 14 (58.33) | 62.5 (31–84)* | 7 (29.17)† | ST 18 (75) | ≥2 cm | 26 | NSCLC 12 (46.15) | 20 (76.92) | 2 |
| Ginalis et al., 202016 | Retro, SC, SA | 12 | 2 (16.67) | 59 (29–83) | 5 (41.67)† | RS 1 (8.33), WBRT 1 (8.33) | >4 cm | 23 | Breast 6 (26.09) | 16 (69.57) | 2 |
| Ito et al., 202010 | Retro, MC, SA | 178 | 98 (55.06) | Breast cancer: 63 (29–90); GI cancer: 69 (35–90); lung: 69 (40–86) | NR | None | >10 cm3 | 182 | NR | NR | 2 |
| Lovo et al., 201931 | Retro, SC, SA | 10 | 2 (20) | 62 (37–77) | 9 (90)‡ | None | Undefined | 22 | Breast 7 (31.82) | 12 (54) | 2 |
| Serizawa et al., 20197 | Retro, MC, CA | 335 | 199 (59.4) | 61 ± 11.7§ | 111 (33.13)‡ | None | >3 cm or >10 cm3 | 485 | NSCLC 14 (2.89) | NR | 2 vs 3 |
| Angelov et al., 201813 | Retro, SC, SA | 54 | 25(46.3) | 63 (23–83) | 16 (29.63)‡ | ST 35 (64.81) | ≥2 cm | 63 | NSCLC 23 (36.51) | 50 (79.37) | 2 |
| Dohm et al., 201811 | Retro, SC, SA | 33 | 22(66.7) | 61 (35.3–85.7) | 9 (27.27)‡ | WBRT 3 (9.09) | Undefined | 39 | Lung 14 (35.9) | NR | 2 |
| Dohm et al., 201812 | Retro, SC, CA | 38 | 16 (42.11) | 62 (35–87) | 14 (36.84)‡ | WBRT 3 (7.89) | ≥6 cm3 | 45 | Lung 17 (37.78) | 34 (75.56) | 2 |
| Yamamoto et al., 201832 | Retro, MC, SA | 78 | 44 (56.41) | 65 (38–86) | 32 (41.03)‡ | RS 5 (6.41) | >10 cm3 | NR | Lung 37 (47.4) | NR | 3 |
| Hasegawa et al., 201730 | Retro, SC, CA | 56 | 37 (66.07) | 69.5 (23–88) | 25(44.64)‡ | OR 4 (7.14) | ≥10 cm3 | 65 | Lung 27 (41.54) | 53 (81.54) | 2 vs 3 |
| Yomo & Hayash, 20148 | Pro, SC, SA | 58 | 37 (63.79) | 66 (32–88) | 70 (30–100)¶ | ST 21 (36.21) | ≥10 cm3 | 61 | Lung 34 (55.74) | 41 (67.21) | 2 |
| Yomo et al., 20129 | Pro, SC, SA | 27 | 18 (66.67) | 65 (32–88) | 60 (30–90)¶ | RS 6 (22.22) | Infratentorial region: >10 cm3; supratentorial region: >15 cm3 | 28 | Lung 17 (60.71) | 13 (46.43) | 2 |
| Higuchi et al., 200934 | Retro, SC, SA | 43 | 24(55.81) | 64 (41–84) | NR | None | ≥10 cm3 | 46 | Colon 14 (30.43) | NR | 3 |
CA = comparative arm; diam = diameter; GI = gastrointestinal; MC = multicenter; NR = not reported; NSCLC = non–small cell lung cancer; OR = Ommaya reservoir; Pro = prospective; pts = patients; Retro = retrospective; RS = resection surgery; SA = single arm; SC = single center; ST = systemic therapy; Tx = treatment.
Age at GKRS1.
Eastern Cooperative Oncology Group (ECOG) Performance Status Scale ≥ 2.
Karnofsky Performance Scale score < 80.
Mean ± SD.
Values expressed as the median Karnofsky Performance Scale score (range).
Risk of Bias Assessment
Ten cohort studies and 2 nonrandomized clinical trials were evaluated using the ROBINS-I tool. The overall risk of bias was serious in 58%7,9,10,14,16,30,34and moderate in 42%.8,11–13,32None of them presented a low risk of bias due to the methodological disadvantages and low-quality evidence provided by observational studies. With respect to the components of the risk of bias assessment tool, 58% of studies7,9,10,14,16,30,34had a serious risk due to confounding bias; 1 study7presented a moderate risk of bias in the classification of interventions; 1 study30had moderate risk of bias due to missing data; 5 studies7,10–12,32had moderate risk of bias in selection of results; and all of the studies (100%) presented an absence of bias in the domains of participants’ selection and deviation from intended interventions, as well as a moderate risk of bias in measurement of outcomes (Supplementary Fig. 1). Only 2 case series31,33were assessed using the Joanna Briggs Institute checklist, and it outlined the absence of clear inclusion criteria and site demographic information (Supplementary Fig. 2).
Volumetric and Survival Data
With respect to the volumetric analysis, volume measurements reported as median and range were manually converted to mean and standard deviation. Three studies did not report disaggregated data per intervention.10,30,32Only 1 study reported the absolute volume reduction: 5.58 ± 7.25 cm3.13The mean LBM volume at GKRS1 ranged from 9.88 to 21.23 cm3. Five studies reported the mean LBM volume at last follow-up.7,13,14,31,34The median overall survival ranged from 6.6 to 24 months after 2-stage GKRS and from 8.3 to 15.9 months after 3-stage GKRS.7,32,34Consistently, the 2-stage GKRS group reported the lowest median progression-free survival as 5.07 (range 3.53–5.23) months16(Supplementary Table 3).
Synthesis of Results
We found a pooled mean volume difference for GKRS1 volume reduction of 4.84 cm3(95% CI 3.80–5.88 cm3), GKRS2 volume reduction of 3.77 cm3(95% CI 1.14–6.40 cm3) and absolute total volume reduction of 6.27 cm3(95% CI 5.67–6.88 cm3) after 2-stage GKRS, consistent with a relative total volume reduction of 54.36% (95% CI 39.92%–68.79%) (Fig. 2). Heterogeneity among studies was high (I293%–96%, p < 0.01). Most studies showed overlapping confidence intervals (Fig. 3).
Relative BM volume reduction from baseline to last follow-up (LFU) in included studies.
Meta-analysis of the mean difference of LBMs’ volume changes.
The effectiveness outcome rate was 44.63% (95% CI 26.50%–64.31%; I287%). The pooled rate of neurological mortality was 16.19% (95% CI 7.68%–30.98%; I280%). All-cause mortality evidenced a pooled rate of 47.92% (95% CI 28.04%–68.49%; I289%) after 2-stage GKRS (Fig. 4).
Proportions meta-analysis of 2-stage GKRS.
Additional Analysis
Following the Cochrane Handbook for Systematic Reviews of Interventions,28most subgroup analyses were only performed for observational studies to avoid mixing them with clinical trials. The complete response rate was significantly higher in cohort (53.26%; 95% CI 35.71%–70.42%) and multicenter (67.03%; 95% CI 59.69%–73.81%) studies, as well as in patients who predominantly had WBRT as prior treatment (65.22%; 95% CI 42.73%–83.62%). Regarding safety outcomes, neurological mortality achieved its highest rates in those who underwent WBRT as prior treatment (28.18%; 95% CI 13.96%–44.78%). All-cause mortality rates were mostly reported in nonrandomized clinical trials (75.84%; 95% CI 65.96%–84.56%), prospective (75.84%; 95% CI 65.96%–84.56%) and multicenter (79.21%; 95% CI 72.51%–84.92%) studies, as well as in patients who showed lung cancer as primary tumor (41.37%; 95% CI 32.19%–50.84%) and who did not undergo any prior treatment (78.12%; 95% CI 71.54%–84.13%) (Table 2).
Subgroup analysis of complete response, neurological mortality, and all-cause mortality of LBMs treated with 2-stage GKRS
| Subgroup | Complete Response | Neurological Mortality | All-Cause Mortality | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| No. | Proportion | 95% CI | % wt | I2 | No. | Proportion | 95% CI | % wt | I2 | No. | Proportion | 95% CI | % wt | I2 | |
| Study design | |||||||||||||||
| NRCT | 0 | NA | NA | NA | NA | 0 | NA | NA | NA | NA | 2 | 75.84 | 65.96–84.56 | 26.32 | NA |
| Cohort | 4 | 53.26 | 35.71–70.42 | 81.28 | 84.82 | 5 | 17.25 | 6.58–31.14 | 88.77 | 83.30 | 4 | 50.68 | 28.14–73.08 | 63.93 | 91.30 |
| Case series | 1 | 9.09 | 1.12–29.16 | 18.72 | NA | 1 | 10.00 | 0.25–44.50 | 11.23 | NA | 1 | 30.00 | 6.67–65.25 | 9.75 | NA |
| Type of recollection | |||||||||||||||
| Retro | 6 | 43.58 | 22.93–65.40 | 100 | 91.18 | 6 | 16.28 | 6.61–28.66 | 100 | 79.15 | 5 | 47.86 | 26.84–69.26 | 73.68 | 90.12 |
| Pro* | 0 | NA | NA | NA | NA | 0 | NA | NA | NA | NA | 2 | 75.84 | 65.96–84.56 | 26.32 | NA |
| No. of centers | |||||||||||||||
| SC | 5 | 36.72 | 14.10–62.71 | 77.91 | 87.91 | 5 | 17.87 | 4.83–35.67 | 78.14 | 80.04 | 4 | 41.15 | 32.58–49.98 | 80.82 | 0.00 |
| MC | 1 | 67.03 | 59.69–73.81 | 22.09 | NA | 1 | 11.24 | 7.00–16.82 | 21.86 | NA | 1 | 79.21 | 72.51–84.92 | 19.18 | NA |
| Predominant sex | |||||||||||||||
| Male | 3 | 61.78 | 54.95–68.39 | 41.32 | NA | 2 | 13.38 | 8.98–18.45 | 39.27 | NA | 2 | 51.27 | 17.69–82.24 | 53.28 | NA |
| Female | 3 | 41.39 | 12.62–73.61 | 58.68 | NA | 4 | 14.67 | 1.13–36.48 | 60.73 | 82.30 | 3 | 45.94 | 34.32–57.77 | 46.72 | NA |
| Predominant primary tumor | |||||||||||||||
| Lung cancer | 3 | 44.59 | 34.24–55.18 | 40.32 | NA | 3 | 22.77 | 4.31–49.11 | 54.66 | NA | 2 | 41.37 | 32.19–50.84 | 52.30 | NA |
| Breast cancer | 2 | 35.09 | 21.47–49.96 | 37.59 | NA | 2 | 9.04 | 0.01–26.39 | 23.47 | NA | 2 | 40.76 | 20.26–62.82 | 28.52 | NA |
| NR | 1 | 67.03 | 59.69–73.81 | 22.09 | NA | 1 | 11.24 | 7.00–16.82 | 21.86 | NA | 1 | 79.21 | 72.51–84.92 | 19.18 | NA |
| Prior Tx | |||||||||||||||
| Yes | 4 | 47.24 | 25.31–69.71 | 59.18 | 0.00 | 4 | 19.47 | 4.43–40.57 | 66.91 | 84.80 | 3 | 42.13 | 33.26–51.24 | 66.90 | NA |
| No | 2 | 60.62 | 53.68–67.36 | 40.82 | NA | 2 | 9.65 | 5.36–14.79 | 33.09 | NA | 2 | 78.12 | 71.54–84.13 | 33.10 | NA |
| Predominant prior Tx | |||||||||||||||
| ST | 2 | 44.59 | 34.24–55.18 | 40.32 | NA | 1 | 5.56 | 1.16–15.39 | 19.27 | NA | 2 | 43.52 | 32.51–54.85 | 34.90 | NA |
| WBRT | 2 | 65.22 | 42.73–83.62 | 18.86 | NA | 3 | 28.18 | 13.96–44.78 | 47.64 | NA | 1 | 39.77 | 25.45–54.94 | 32 | NA |
| None | 2 | 60.62 | 53.68–67.36 | 40.82 | NA | 2 | 9.65 | 5.36–14.79 | 33.09 | NA | 2 | 78.12 | 71.54–84.13 | 33.10 | NA |
NA = not applicable; NRCT = nonrandomized clinical trial; wt = weight.
Pooled estimate from experimental studies.
Evidence Certainty
Regarding the certainty of evidence for complete response, neurological mortality, and GKRS2 volume reduction, we started the assessment with the assumption of low quality because most studies were not comparative. We downgraded by 1 point because almost all studies had a serious risk of bias and indirectness in their population. Neurological mortality was downgraded 2 points because of the heterogeneity among studies and obtained an overall very low certainty.
全因死亡率得到很低的确定性for the observational studies, but a moderate certainty for the experimental. In the first instance, we started the assessment with the assumption of low quality and reduced the rating by 1 point because most of the individual studies had a serious risk of bias, and we also reduced the certainty by 1 point because the cutoff volume or diameter used for the definition of LBMs substantially differed among studies (Table 3).
Summary of findings in patients who underwent staged GKRS for LBMs
| Outcome | No. of Participants (no. & type of studies) | GRADE Certainty Assessment | Effect (95% CI) | Certainty of Evidence* | ||||
|---|---|---|---|---|---|---|---|---|
| Risk of Bias | Indirectness | Inconsistency | Imprecision | Publication Bias | ||||
| Total vol reduction | 470†(5 obs) | Serious‡ | Serious§ | Not serious | Not serious | NA | MD 6.27 cm3(5.67–6.88 cm3) | Low |
| GKRS1 vol reduction | 220†(6 obs) | Serious‡ | Serious§ | Not serious | Not serious | NA | MD 4.84 cm3(3.80–5.88 cm3) | Low |
| GKRS2 vol reduction | 134†(4 obs) | Serious‡ | Serious§ | Not serious | Very serious¶ | NA | MD 3.77 cm3(1.14–6.40 cm3) | Very low |
| Complete response | 314†(5 obs) | Serious‡ | Serious§ | Very serious** | Not serious | NA | 44.63% (26.50–64.31%) | Very low |
| Neurological mortality | 325 (6 obs) | Serious‡ | Very serious§ | Very serious** | Not serious | NA | 16.19% (7.68–30.98%) | Very low |
| All-cause mortality (obs) | 311 (6 obs) | Serious‡ | Very serious§ | Very serious** | Not serious | NA | 47.92% (28.04–68.49%) | Very low |
| All-cause mortality (exp) | 85 (2 exp) | Serious‡ | Not serious | Not serious | Not serious | NA | 75.84% (65.96–84.56%) | Moderate |
Exp = experimental; GRADE = Grading of Recommendations Assessment, Development, and Evaluation; MD = mean difference; obs = observational.
The GRADE Working Group grades of evidence were defined as follows: high certainty, we are very confident that the true effect lies close to that of the estimate of the effect; moderate certainty, we are moderately confident in the effect estimate—the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different; low certainty, our confidence in the effect estimate is limited—the true effect may be substantially different from the estimate of the effect; and very low certainty, we have very little confidence in the effect estimate—the true effect is likely to be substantially different from the estimate of the effect.
Number of brain metastases.
Less than 25% of studies have low risk of bias.
Indirectness: populations are substantially different regarding LBM cutoff criteria or predominant tumor foci.
Imprecision due to large confidence interval.
Inconsistency: estimates are heterogeneous in > 75% of studies.
Discussion
Our systematic review highlights the features of 2-stage GKRS by representing the main clinical outcomes (volumetric outcomes, all-cause mortality, neurological mortality, and complete response) via synthetizing effect measures. Fourteen studies conducted between 2009 and 2022 were analyzed in the review criteria for the use of a 2-stage GKRS protocol for LBMs. A total of 1019 brain metastases in 958 patients were included. The meta-analysis shows a significant volume reduction of 54.36% for the LBMs, with an absolute reduction on average of 6 cm3at the completion of the staged GKRS. Complete response was seen in 44.6% of the pooled patients. Meanwhile, the neurological morbidity from progression was 16.19% and all-cause mortality was 47.92%. Overall, there was a serious risk of bias across the studies. The absolute volume reduction was similar to hypofractionated GKRS, but the average percent change was lower with staged GKRS; 54.36% reduction as compared to up to a 90% reduction from baseline volume.35,36Considering that a preoperative diameter of > 3 cm is a clinical predictor of local failure,37,38effectively reducing mass effect with staged GKRS provides a nonoperative option to prevent acute morbidity from compression of eloquent brain tissue.
For clinical outcomes, the 2-stage GKRS showed a comparable if not superior complete response to all currently reported hypofractionated GKRS, which ranged from 15%–43% in current literature.39–41正如所料,完整的响应se rate for staged GKRS was barely half of the control rate reported for postresection directed GKRS, and thus, staged GKRS is not a replacement for surgical cytoreduction. Evidence reported in the literature shows LBMs to be highly related to neurological and all-cause mortality, given the development of intracranial hypertension in the postoperative course. Staged GKRS had a lower but not statistically significant neurological mortality rate (including radiation necrosis, neoplasm expansion, edema-related intracranial hypertension, and others) compared to hypofractionated GKRS, in which adverse radiation effects still constitute 24% of morbidity rates.42,43On the other hand, rates of procedure-related mortality are lower compared to those for resection, given the absence of brain tissue manipulation that injures white matter tracts and functional cortex.44,45Despite the seemingly improved clinical outcomes, potential downsides of staged SRS include dynamic changes in radiation dose, which might be a cause of incomplete treatment,46as well as the absence of histopathological evaluation between stages.
The certainty of the evidence within the meta-analysis was predominantly very low. The outcomes were downgraded given the indirectness among studies, relying on the heterogeneous cutoff diameter used for classifying LBMs, primary tumor foci, and high imprecision, which are expressed in the large confidence intervals. Moreover, the risk of inconsistency was high because many outcomes reported an I2> 75%. Most studies were observational, retrospective, and single-center reports, including cohorts7,10–14,16,30,32,34and case series.31,33Only 2 were nonrandomized clinical trials.8,9This implies some serious methodological drawbacks, given the absence of control of confounders, leading to a decrease in internal validity of results and quality of evidence. Thus, we recommend that further experimental studies be performed to achieve consensus on the use of 2-stage GKRS for LBMs.
Limitations and Strengths
The systematic review encountered limitations anticipated for a novel therapy in its nascence. Of a total of 14 included studies, only 2 were prospective experimental designs. In addition, there was variability within the treatment protocol; 3 of the 14 studies included patients with a 3-stage GKRS protocol. Only 2 addressed a direct comparison between the 2-stage and the 3-stage GKRS for LBM, and thus a comparative meta-analysis between the 2 protocols could not be performed. Moreover, the cutoff size (either volume or maximum diameter) criteria for defining an LBM was highly heterogeneous among studies, leading to a decreased internal validity of the results obtained through the meta-analysis. Despite this, patient enrollment was predominantly guided by the maximum diameter cutoff as inclusion criteria, leading to a broad range in BM volumes (9.88–21.33 cm3) and a size gap below this value. Finally, given the absence of data on follow-up time, the absolute volume reduction after GKRS could not be normalized to a standardized mean follow-up duration among studies. Most of these limitations were addressed by running a subgroup analysis to account for the main confounders while assessing the 2-stage GKRS clinical outcomes.
Conclusions
This systematic review provides the largest available analysis of staged GKRS, with 1019 total metastases. In the meta-analysis, the 2-stage GKRS achieved a mean 54.36% or > 6-cm3reduction in total volume, as well as a 44% complete response rate, revealing effectiveness and safety for the treatment of LBMs. This is still a novel technique, and weak methodological approaches and the lack of robust literature limits the results to a low to very low certainty of evidence. Therefore, although promising, staged GKRS requires higher-quality primary studies prior to universal recommendation for LBMs.
Disclosures
The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper.
Author Contributions
概念和设计:特里,Merenzon Daggubati,Luther, Shah. Acquisition of data: Terry, Zullo, Komotar. Analysis and interpretation of data: Terry, Merenzon, Zullo, Bhatia, Ivan. Drafting the article: Terry, Levy. Critically revising the article: Terry, Merenzon, Daggubati, Levy, Bhatia, Luther, Ivan. Reviewed submitted version of manuscript: Terry, Merenzon, Daggubati, Levy, Bhatia. Approved the final version of the manuscript on behalf of all authors: Terry. Statistical analysis: Terry, Bhatia. Administrative/technical/material support: Daggubati, Zullo. Study supervision: Terry, Merenzon, Daggubati, Shah.
Supplemental Information
Online-Only Content
Supplemental material is available online.
Supplementary Tables and Figures.//www.prize-show.com/doi/suppl/10.3171/2023.5.FOCUS23232.
References
-
2 ↑
ProescholdtMA,SchödelP,DoenitzC,et al.The management of brain metastases—systematic review of neurosurgical aspects.Cancers (Basel).2021;13(7):1616.
-
3 ↑
SarmeyN,Kaisman-ElbazT,MohammadiAM.Management strategies for large brain metastases.Front Oncol.2022;12:827304.
-
4 ↑
MinnitiG,ScaringiC,PaoliniS,et al.Single-fraction versus multifraction (3 × 9 Gy) stereotactic radiosurgery for large (>2 cm) brain metastases: a comparative analysis of local control and risk of radiation-induced brain necrosis.Int J Radiat Oncol Biol Phys.2016;95(4):1142–1148.
-
5
ManningMA,CardinaleRM,BenedictSH,et al.Hypofractionated stereotactic radiotherapy as an alternative to radiosurgery for the treatment of patients with brain metastases.Int J Radiat Oncol Biol Phys.2000;47(3):603–608.
-
6 ↑
WegnerRE,LeemanJE,KabolizadehP,et al.Fractionated stereotactic radiosurgery for large brain metastases.Am J Clin Oncol.2015;38(2):135–139.
-
7 ↑
SerizawaT,HiguchiY,YamamotoM,et al.Comparison of treatment results between 3- and 2-stage Gamma Knife radiosurgery for large brain metastases: a retrospective multi-institutional study.J Neurosurg.2019;131(1):227–237.
-
8 ↑
YomoS,HayashiM.A minimally invasive treatment option for large metastatic brain tumors: long-term results of two-session Gamma Knife stereotactic radiosurgery.Radiat Oncol.2014;9:132.
-
9 ↑
YomoS,HayashiM,NicholsonC.A prospective pilot study of two-session Gamma Knife surgery for large metastatic brain tumors.J Neurooncol.2012;109(1):159–165.
-
10 ↑
ItoD,AoyagiK,NaganoO,SerizawaT,IwadateY,HiguchiY.Comparison of two-stage Gamma Knife radiosurgery outcomes for large brain metastases among primary cancers.J Neurooncol.2020;147(1):237–246.
-
11 ↑
DohmA,McTyreER,OkoukoniC,et al.Staged stereotactic radiosurgery for large brain metastases: local control and clinical outcomes of a one-two punch technique.开云体育app官方网站下载入口.2018;83(1):114–121.
-
12 ↑
DohmAE,HughesR,WhelessW,et al.Surgical resection and postoperative radiosurgery versus staged radiosurgery for large brain metastases.J Neurooncol.2018;140(3):749–756.
-
13 ↑
AngelovL,MohammadiAM,BennettEE,et al.Impact of 2-staged stereotactic radiosurgery for treatment of brain metastases ≥ 2 cm.J Neurosurg.2018;129(2):366–382.
-
14 ↑
DamronEP,DonoA,ChafiH,et al.Metastatic neoplasm volume kinetics following 2-stage stereotactic radiosurgery.World Neurosurg.2022;161:e210–e219.
-
15
FrischerJM,FrallerA,MallouhiA,et al.Evaluation of dose-staged Gamma Knife radiosurgical treatment method for high-risk brain metastases.World Neurosurg.2016;94:352–359.
-
16 ↑
GinalisEE,CuiT,WeinerJ,NieK,DanishS.两步立体定向放射外科治疗ment of large brain metastases: single institution experience and review of literature.J Radiosurg SBRT.2020;7(2):105–114.
-
17 ↑
VogelbaumMA,BrownPD,MessersmithH,et al.Treatment for brain metastases: ASCO-SNO-ASTRO guideline.J Clin Oncol.2022;40(5):492–516.
-
18 ↑
PageMJ,McKenzieJE,BossuytPM,et al.The PRISMA 2020 statement: an updated guideline for reporting systematic reviews.BMJ.2021;372:n71.
-
19 ↑
Terry EscalanteFA,MerenzonMA,IvanME,KomotarRJ.Time-staged stereotactic radiosurgery for large brain metastases: a systematic review and metanalysis of clinical outcomes.普洛斯彼罗国际前瞻性登记系统tematic Reviews.Published online March 9, 2023. Accessed June 14, 2023.https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=384463
-
21 ↑
SterneJA,HernánMA,ReevesBC,et al.ROBINS-I: a tool for assessing risk of bias in non-randomised studies of interventions.BMJ.2016;355:i4919.
-
22 ↑
MunnZ,BarkerTH,MoolaS,et al.Methodological quality of case series studies: an introduction to the JBI critical appraisal tool.JBI Evid Synth.2020;18(10):2127–2133.
-
23 ↑
LinNU,LeeEQ,AoyamaH,et al.Response assessment criteria for brain metastases: proposal from the RANO group.Lancet Oncol.2015;16(6):e270–e278.
-
25 ↑
FreemanMF,TukeyJW.Transformations related to the angular and the square root.Ann Math Statist.1950;21(4):607–611.
-
26 ↑
HozoSP,DjulbegovicB,HozoI.Estimating the mean and variance from the median, range, and the size of a sample.BMC Med Res Methodol.2005;5(1):13.
-
27 ↑
DaltonJE,BolenSD,MaschaEJ.Publication bias: the elephant in the review.Anesth Analg.2016;123(4):812–813.
-
28 ↑
HigginsJPT,ThomasJ.ChandlerJ,et al., eds.科克伦手册系统atic Reviews of Interventions. Version 6.3.Cochrane;2022.Accessed January 24, 2023.https://training.cochrane.org/handbook
-
29 ↑
SchünemannH,BrożekJ,GuyattG.OxmanA, eds.GRADE Handbook.Updated October 2013. Accessed June 14, 2023.https://gdt.gradepro.org/app/handbook/handbook.html
-
30 ↑
HasegawaT,KatoT,YamamotoT,et al.Multisession gamma knife surgery for large brain metastases.J Neurooncol.2017;131(3):517–524.
-
31 ↑
LovoEE,TorresLB,CamposFJ,et al.Two-session radiosurgery as initial treatment for newly diagnosed large, symptomatic brain metastases from breast and lung histology.Cureus.2019;11(8):e5472.
-
32 ↑
YamamotoM,HiguchiY,SerizawaT,et al.Three-stage Gamma Knife treatment for metastatic brain tumors larger than 10 cm3: a 2-institute study including re-analyses of earlier results using competing risk analysis.J Neurosurg.2018;129(suppl 1):77-85.
-
33 ↑
CuiT,WeinerJ,DanishS,et al.Evaluation of biological effective dose in Gamma Knife staged stereotactic radiosurgery for large brain metastases.Front Oncol.2022;12:892139.
-
34 ↑
HiguchiY,SerizawaT,NaganoO,et al.Three-staged stereotactic radiotherapy without whole brain irradiation for large metastatic brain tumors.Int J Radiat Oncol Biol Phys.2009;74(5):1543–1548.
-
35 ↑
MoonHC,ParkYS.Volume prediction for large brain metastases after hypofractionated gamma knife radiosurgery through artificial neural network.Medicine (Baltimore).2022;101(40):e30964.
-
36 ↑
ParkYG,KimEY,ChangJW,ChungSS.Volume changes following gamma knife radiosurgery of intracranial tumors.Surg Neurol.1997;48(5):488–493.
-
37 ↑
JagannathanJ,YenCP,RayDK,et al.Gamma Knife radiosurgery to the surgical cavity following resection of brain metastases.J Neurosurg.2009;111(3):431–438.
-
38 ↑
BrennanC,YangTJ,HildenP,et al.A phase 2 trial of stereotactic radiosurgery boost after surgical resection for brain metastases.Int J Radiat Oncol Biol Phys.2014;88(1):130–136.
-
39 ↑
FahrigA,GanslandtO,LambrechtU,et al.Hypofractionated stereotactic radiotherapy for brain metastases—results from three different dose concepts.Strahlenther Onkol.2007;183(11):625–630.
-
40
NarayanVM,KonetyBR,WarlickC.Novel biomarkers for prostate cancer: an evidence-based review for use in clinical practice.Int J Urol.2017;24(5):352–360.
-
41 ↑
AndersonES,PostowMA,WolchokJD,et al.Melanoma brain metastases treated with stereotactic radiosurgery and concurrent pembrolizumab display marked regression; efficacy and safety of combined treatment.J Immunother Cancer.2017;5(1):76.
-
42 ↑
MilanoMT,GrimmJ,NiemierkoA,et al.Single- and multifraction stereotactic radiosurgery dose/volume tolerances of the brain.Int J Radiat Oncol Biol Phys.2021;110(1):68–86.
-
43 ↑
ChenWC,BaalUH,BaalJD,et al.Efficacy and safety of stereotactic radiosurgery for brainstem metastases: a systematic review and meta-analysis.JAMA Oncol.2021;7(7):1033–1040.
-
44 ↑
PalmiscianoP,FeriniG,KhanR,et al.Neoadjuvant stereotactic radiotherapy for brain metastases: systematic review and meta-analysis of the literature and ongoing clinical trials.Cancers (Basel).2022;14(17):4328.
-
45 ↑
PatelAJ,SukiD,HatibogluMA,et al.Factors influencing the risk of local recurrence after resection of a single brain metastasis.J Neurosurg.2010;113(2):181–189.
-
46 ↑
UdovicichC,PhillipsC,KokDL,et al.Neoadjuvant stereotactic radiosurgery: a further evolution in the management of brain metastases.Curr Oncol Rep.2019;21(8):73.
