Multi-institutional study of ‘Sandwich treatment’ for motor area large brain metastases (LBM) with diameter over 3 cm

Brain metastases (BM) is the most common intracranial malignant tumor in adults, and is also the main cause of mortality of cancer patients.¹ The current guidelines suggest that patient with limited number of BM with good performance can be treated with stereotactic radiosurgery (SRS) alone.^2,3,4,5 For patients with BM number less than 10, or the total volume smaller than 4 cm³, fractioned radiosurgery or two-staged stereotactic radiosurgery (2-SSRS) could control the tumor growth with low neurotoxicity and not delay the consequent systemic treatment.^{6,7,8,9,10,11}

However, for large BM (LBM) with diameter over 3 cm (minimum volume >= 15 cm³) and located in motor area, even 2-SSRS is still challenging.¹² The risk mainly comes from the compression of LBM and consequent peri-tumor edema to the brain, results in devastating intracranial hypertension. Further, the SRS-induced edema would add to the risk of intracranial hypertension, limb hemiplegia and refractory epilepsy.^13,14 Meanwhile, SRS would cause the brain radiation necrosis (RN).^15,16,17 These potential risks make it difficult for patients with LBM to receive 2-SSRS in outpatient department. Patient after SRS needs long-term inpatient steroids therapy to control peri-tumor and SRS-induced edema to improve their symptoms, and lower the potential risk of RN. Bevacizumab, an anti-VEGF monoclonal molecular drug, has been utilized by practitioners in anti-tumor therapy for cancer patients.^18,19 Also, its anti-angiogenesis effect could be used for SRS-edema and RN control.^20,21 To help patient with motor area LBM and shorten their treatment course, the practitioners from four gamma knife center developed the ‘Sandwich treatment’ strategy. The strategy was one-week treatment course that includes 2-SSRS and using bevacizumab once during SRS gap. In the present study, the authors retrospectively reviewed the patients that received ‘Sandwich treatment’. The purpose of this study was to evaluate the efficacy of and safety of this strategy.

In recent years, SRS has been recognized as effective alternative treatment for BM.^2,3,4 For BM with diameter at 2–3 cm, fractionated or staged SRS is preferable.^9,11,29 One of the representative study on 2-SSRS method for BM is reported by Angelov et al. in 2018.¹⁰ In their case series, the volume of 63 BMs in 54 patients reduced significantly, and the TLC rate in 3, 6 months after treatment reaches 95% and 88%, respectively; the incidence of overall radiation side effects was 11%. Another representative study was conducted by Serizawa T et al.¹¹ They compared the treatment results between 3- and 2-stage Gamma Knife radiosurgery for large BM and found no differences between in terms of patients’ overall survival, tumor progression, neurological death, and radiation-related adverse events. Dohm Amoren et al. reported 2-SRSS for BM that are difficult to be removed by surgery⁸, the cumulative incidence of local treatment failure at 6 and 12 months was 3.2% and 13.3% respectively. The study of Damron et al. in 2022 also supports the effectiveness and safety of 2-SSRS for BM patients.³⁰ According to the results of these studies, 2-SSRS for the BM treatment is satisfactory and local control failure rate is low. Table 3 listed 8 studies that adopted 2-SSRS strategy for BM.

The listed studies mainly focused on the optimal prescription doses of the 2-SSRS. The researchers suggested dose reduction strategy was suitable for lessening of SRS-induced edema and RN. However, the 2-SSRS strategy might not be enough for the LBM located in motor area. Meanwhile, these conventional 2-SSRS strategy have treatment course longer than 1 month. Patients may have to receive even longer time of inpatient dehydration and steroid treatment during the 2-SSRS when the diameter of BM was over 3 cm and located in the motor area. The long-term usage of mannitol and steroid hormones would bring a series of side effects to the patient, delay their consequent systemic treatment, increase their treatment cost. On the other hand, patient’s intracranial hypertension, neuro-dysfunction symptoms and the risk of suffering from refractory epilepsy would make them hardly to receive SRS at outpatient department.

The side effect of 2-SSRS should not be neglected as well. The SRS-induced edema would aggravate the edema caused by LBM compression in the motor area, patients would experience deterioration of limb movement and their life quality. Cho et al. pointed out even 2-SSRS strategy for BM could cause RN as high as 17% and affects the life quality of patients.³³ Considering these tough issues, we introduce bevacizumab. Juan et al. had applied bevacizumab for the treatment of SRS related edema, and found the edema volume reduction at 49.0%–66.0% on MRI-T2FLAIR.³⁴ As bevacizumab could reduce the angiogenesis around the lesion, it could also lower the brain RN risk.³⁵ Its anti-angiogenesis effect could be used to control tumor growth had synergies with SRS. Therefore, the ‘Sandwich treatment’ strategy would have obvious advantage in the treatment of LBM in motor area, especially for those with primary lung adenocarcinoma.³⁶

The results of this retrospective study confirmed the effectiveness and safety of ‘Sandwich treatment’. The results indicated that using bevacizumab during SRS gap could reduce the median volume of peri-tumor edema by 38.8% (from 12.6 cm³ to 5.2 cm³). Meanwhile, patient’s muscle strength score that reflecting the patients’ physical activities also significantly improved. The strategy would shorten the whole treatment course while preserve patients’ neuro-function or improve their life quality. Patients with primary lung adenocarcinoma had significant lower risk of tumor relapse. Compared to previous 2-SSRS reports as showed in Table 3, the present study had the largest median tumor volume, shortest treatment course, while the TLC were similar compared to other studies. The 6-month OS rate and median survival in our present study exceeded 6 of the 8 studies.

The incidence of RN in this study was higher than other eight 2-SSRS cohorts listed in Table 3. The main reason may be that in these cohort the patient’s median OS was short and they failed to report the potential RN symptoms. On contrary, the LBM in the present study were all located in the motor area, and patients’ SRS-induced symptoms could be much more obvious.

As far as we know, this is the first report of ‘Sandwich treatment’ strategy for the LBM with diameter over 3 cm at motor area. The statistical results verified the effectiveness and safety of this strategy. This strategy could significantly improve patients’ life quality and greatly shortened treatment interval. However, a larger cohort is still needed for prospective study.