Re-irradiation followed by resection for recurrent brain metastases after initial stereotactic radiosurgery: illustrative cases

Abstract

BACKGROUND

The management of recurrent brain metastases after initial stereotactic radiosurgery (SRS) remains challenging, with high local recurrence rates following salvage surgery. While preoperative SRS has shown promise for newly diagnosed metastases, its application in post-SRS recurrences is largely unexplored.

OBSERVATIONS

The authors treated three patients with recurrent brain metastases using preoperative re-irradiation followed by resection. Gamma Knife SRS was performed at marginal doses of 16–18 Gy, followed by resection within 24–48 hours. Local control was achieved in all patients without radiation-related complications. Histopathological examination confirmed the presence of viable tumor cells and radiation-induced changes. Follow-up magnetic resonance imaging revealed no evidence of tumor recurrence or adverse effects. Two patients remained alive at 35 and 19 months, whereas one died of primary cancer progression at 20 months.

LESSONS

Preoperative re-irradiation followed by resection for recurrent brain metastases after initial SRS is feasible and offers promise for short-term safety, local control, and rapid symptom improvement. The ability to promptly implement SRS enables its application in oncological emergencies. These findings suggest that preoperative re-irradiation can be a valuable strategy for managing symptomatic post-SRS recurrent brain metastases that require prompt surgical intervention.

https://thejns.org/doi/10.3171/CASE24737

The management of brain metastases has significantly evolved in recent years, with stereotactic radiosurgery (SRS) being increasingly used as an alternative to whole-brain radiation therapy due to its superior local control and preservation of neurocognitive function.^1–4 While postoperative SRS has become standard practice for surgically resected brain metastases and improved local control rates, it faces challenges, including an increased risk of leptomeningeal dissemination and the potential for radiation-induced complications.^1–3,⁵ Additionally, there are technical difficulties in targeting irregular resection cavities and challenges in the accurate assessment of residual tumors via imaging.

To address these challenges, preoperative SRS has emerged as a promising treatment paradigm for newly diagnosed brain metastases.^6–13 Several studies have presented the potential benefits of preoperative SRS, including improved local control, reduced risk of leptomeningeal disease, and decreased radiation exposure to the surrounding normal brain tissue.^14–19 While the safety of repeat SRS for nonsurgical recurrences has been reported,^20–22 the optimal management of recurrent brain metastases that require surgical intervention after initial SRS remains unclear.

In response to the high recurrence rates after surgery for SRS-treated metastases,²³ we developed a novel treatment approach that involves preoperative re-irradiation, followed by resection for recurrent brain metastases after initial SRS treatment. This strategy combines the benefits of re-irradiation and surgical removal while potentially reducing the risk of local recurrence and leptomeningeal spread in this challenging patient population.

In this case series, we report our experience with preoperative repeat SRS in three patients with recurrent brain metastases after prior SRS. We discuss the feasibility, safety, and potential benefits of this innovative approach for the management of post-SRS recurrence. Our findings provide valuable insights into treatment options for this complex clinical scenario and pave the way for further investigation of this strategy.

Illustrative Cases

Patient Selection and Characteristics

At our institution, SRS is the standard initial treatment for brain metastases, except in patients who require immediate decompression. This patient series includes three women (ages 41–65 years) with recurrent brain metastases from different primary cancers (lung adenocarcinoma, endometrial cancer, and breast cancer) who underwent preoperative re-irradiation followed by resection. These patients had a history of SRS treatment, with an initial response followed by tumor regrowth after at least 6 months. In this series, radiation treatments were classified as single-fraction SRS or multisession SRS for hypofractionated treatments. The T1/T2 match criterion was used to differentiate between radiation necrosis and tumor recurrence.²⁴ According to this criterion, correspondence between contrast-enhanced volume on T1-weighted images and low signal–defined lesion margin on T2-weighted images, along with enlargement of the enhancing lesion and mass effect, indicated tumor recurrence. The decision for surgical intervention was made when patients exhibited neurological deterioration due to tumor progression, accompanied by significant peritumoral edema.

Imaging and Follow-Up Protocol

Following initial SRS treatment, all patients underwent regular follow-ups with magnetic resonance imaging (MRI) every 3 months. MRI examinations were performed using a 1.5T scanner (Siemens) with standard brain tumor protocols, including pre- and postcontrast T1-weighted, T2-weighted, fluid-attenuated inversion recovery, and diffusion-weighted imaging sequences. This imaging protocol was maintained throughout the study period, including after preoperative re-irradiation and resection. Regular clinical examinations were conducted in conjunction with MRI. Patients were followed up continuously, with ongoing follow-ups for surviving patients.

Treatment Approach

Re-irradiation was performed using Gamma Knife SRS (Elekta AB) with marginal doses of 16 or 18 Gy at the 60% isodose line, followed by resection within 24–48 hours. The re-irradiation dose was determined based on the cumulative dose from previous treatments and proximity to critical structures. The radiation target was defined as the contrast-enhancing region, excluding the nonenhancing cystic components. Resection was performed via craniotomy with the aid of neuronavigation, aiming for gross-total resection while preserving neurological function.

Histopathological Examination

Histopathological examinations were performed on all resected specimens. Samples were fixed in 10% neutral buffered formalin, embedded in paraffin, and sectioned at 5-μm thickness. Hematoxylin and eosin (H&E) staining was performed to assess general morphology and radiation-induced changes. Immunohistochemical staining for Ki-67 was performed to assess tumor cell proliferation. Additionally, CAM5.2 and CK7 immunostaining were performed to further characterize the tumor cells and assess their origin. Radiation necrosis was evaluated based on the presence of coagulative necrosis, vascular changes, and reactive gliosis.

Case Presentations

The patient and treatment characteristics are summarized in Table 1.

TABLE 1.

Patient and treatment characteristics

Variable	Case 1	Case 2	Case 3
Age (yrs)	42	65	41
Sex	F	F	F
Primary cancer	Lung adenocarcinoma	Endometrial cancer	Breast cancer
Tumor location	Lt occipital	Rt parietal	Rt cerebellum
Prior SRS marginal dose (Gy)	20 (initial), 18 (at 16 mos)	35 in 5 fractions	35 in 5 fractions
Time since last SRS (mos)	15	7	26
Symptom	Rt visual field defect, Gerstmann syndrome, rt upper-limb paresis	Lt hemiparesis, lt sensory loss	Cerebellar ataxia
Tumor vol (cm3)	25.8	17.9	5.4
Preop SRS marginal dose (Gy)	16	18	18
Preop SRS isodose (%)	60	60	60
Follow-up duration (mos)	35	20	19
Local control	No recurrence	No recurrence	No recurrence
Meningeal dissemination	None	None	None
New lesions	None	None	None
Outcome	Alive	Dead (primary cancer)	Alive

Case 1

A 42-year-old woman was initially diagnosed with brain metastasis from lung adenocarcinoma. At presentation, she had multiple small asymptomatic lesions, all of which were treated with SRS (marginal dose of 20 Gy). Sixteen months after the treatment, follow-up imaging revealed regrowth of the occipital lesion; however, the patient remained asymptomatic. Given the lack of symptoms, re-irradiation with SRS (marginal dose of 18 Gy) was chosen over resection (Fig. 1A). However, 15 months after the second SRS, the patient developed significant neurological symptoms, including a right visual field defect, Gerstmann syndrome, and right upper-limb paresis. MRI revealed tumor progression with extensive peritumoral edema (Fig. 1B and C), and the T1/T2 match suggested tumor recurrence rather than radiation necrosis. Based on this assessment, preoperative repeat SRS was performed using a marginal dose of 16 Gy, followed by gross-total resection on the following day. The patient remained recurrence free after 35 months of follow-up (Fig. 1D).

FIG. 1.

Case 1. Right parieto-occipital region (A–D). Case 2. Left parietal lobe (E–H). Case 3. Left cerebellar hemisphere (I–L). Previous SRS (A,E,and I). Preoperative gadolinium-enhanced T1-weighted imaging (GdT1-WI; B, F, and J). Preoperative T2-weighted imaging (T2-WI; C, G, and K). Postoperative MRI (D, H, and L). Preoperative GdT1-WI and T2-WI present a T1/T2 match suggestive of tumor recurrence, with significant peritumoral edema on T2-WI. Postoperative MRI shows no evidence of tumor recurrence or adverse radiation effects in the surrounding brain tissue.

Case 2

A 65-year-old woman diagnosed with brain metastasis from endometrial cancer presented with a mild right-sided sensory disturbance. Because of the large size of the lesion and its proximity to the motor cortex, fractionated stereotactic radiation therapy (FSRT) (marginal dose of 35 Gy in 5 fractions) was administered. The tumor initially responded well, with shrinkage of the lesion and improvement in symptoms. However, 7 months later, the patient developed left hemiparesis and worsened left sensory loss. Although the tumor size was slightly smaller than at the initial presentation, MRI showed enlargement of the cystic component of the tumor and worsening of the peritumoral edema (Fig. 1F and G). The patient underwent resection after receiving preoperative repeat SRS with a marginal dose of 18 Gy. Although the patient died of progression of the primary tumor 20 months after surgery, there was no evidence of brain metastasis (Fig. 1H).

Case 3

A 41-year-old woman was diagnosed with asymptomatic cerebellar metastasis from breast cancer. The patient underwent FSRT (marginal dose of 35 Gy in 5 fractions), which resulted in tumor shrinkage. However, 26 months later, the patient developed severe ataxia, and MRI revealed cystic enlargement of the tumor with peritumoral edema (Fig. 1J and K). Given the significant symptom progression and T1/T2 match indication, preoperative repeat SRS with an 18-Gy marginal dose targeting the enhancing component was performed, followed by resection. The patient remained recurrence free after 19 months of follow-up (Fig. 1L).

Overall Results

In all three patients, the decision for surgical intervention was based on the development or worsening of neurological symptoms coupled with radiological evidence of tumor progression and significant peritumoral edema. No patients experienced transient worsening of their neurological symptoms during the 24- to 48-hour interval between SRS and resection. All patients experienced complete resolution of their presenting symptoms after surgery, with no radiation-related complications. Local tumor control was achieved in all patients, with no evidence of meningeal dissemination or development of new lesions during the follow-up period.

Notably, histopathological examination confirmed a mixture of radiation necrosis and viable tumor cells in all specimens (Fig. 2), thus validating the diagnosis of recurrence rather than pure radiation necrosis. Follow-up MRI scans showed no evidence of tumor recurrence, adverse radiation effects on the surrounding brain tissue, or development of new metastatic lesions that required additional treatment.

FIG. 2.

Case 1. Lung adenocarcinoma (A–C). Case 2. Endometrial cancer (D–F). Case 3. Breast cancer (G–I). H&E staining at low magnification. Bar = 500 μm (A, D, and G). H&E staining at high magnification. Bar = 20 μm (B and E) and 50 μm (H). Immunohistochemical staining for CAM5.2 (C and I) and CK7 (F). Bar = 20 μm (C and F) and 100 μm (I). All specimens show a mixture of viable tumor cells and radiation-induced changes, including vascular wall thickening with fibrinoid degeneration, hyalinization, focal necrosis, areas of coagulative necrosis, and reactive gliosis in the surrounding brain tissue.

Informed Consent

The necessary informed consent was obtained in this study.

Discussion

Observations

The management of adjuvant radiation therapy for surgically treated brain metastases has evolved significantly over the past few decades. This approach has transitioned from postoperative whole-brain radiation therapy to postoperative SRS^1–5 and, more recently, to preoperative SRS for initial treatment.^6–19 This shift was driven by the potential to reduce both the risk of leptomeningeal disease and radiation-induced toxicity.^{6,12,14,15,17,18}

The optimal management of recurrence after initial SRS remains controversial and presents unique challenges. When recurrence is diagnosed, re-irradiation is one option;^{20–22 25} however, in patients with neurological symptoms or significant mass effects, resection may be necessary. Prior studies have reported local recurrence rates of up to 40% following salvage surgery for SRS-treated recurrent brain metastases, even after gross-total resection.²³ While the risk of local recurrence and leptomeningeal spread in patients with recurrence is not necessarily higher than that in newly diagnosed patients, it remains a significant concern.

To address these challenges, our institution has adopted a novel approach of preoperative re-irradiation followed by resection in patients diagnosed with symptomatic recurrence of brain metastases after initial SRS. This strategy mirrors the rationale behind preoperative SRS for newly diagnosed brain metastases.^6–19 By irradiating an intact tumor before resection, we aim to reduce the viability of tumor cells and potentially decrease the risk of local recurrence and leptomeningeal spread.^6–19

In our study, histopathological examination of the resected specimens revealed viable tumor cells mixed with radiation-induced necrosis, which confirms true tumor recurrence.²³,²⁶ This finding underscores the critical concern regarding active tumor cells in post-SRS recurrence, which can lead to local recurrence and leptomeningeal dissemination if not adequately addressed.²³ Importantly, despite the presence of viable tumor cells in the resected specimens, our patients showed no evidence of local recurrence or leptomeningeal spreading during the follow-up period. This observation indicates that our approach of preoperative re-irradiation followed by resection effectively managed tumor recurrence while potentially preventing further spread.

However, this approach of re-irradiation for recurrent brain metastases necessitates careful consideration of potential risks, particularly the increased likelihood of radiation necrosis.^20–22,25 Prior studies have shown that the 1-year cumulative incidence of adverse radiation effects after repeat SRS is significantly higher (20%) than the incidence after the initial SRS (5%).²⁶ Despite these concerns, when carefully selected and planned, re-irradiation can provide effective local control with manageable toxicity.^20–22 In our approach, we employed relatively high doses for preoperative re-irradiation, with marginal doses of 18 Gy for the second irradiation session and 16 Gy for the third irradiation session,^20–22,25 and our patients did not experience significant adverse radiation effects during the follow-up period. Furthermore, our follow-up period, which ranged from 19 to 35 months, was relatively short for long-term toxicity assessment but extended well beyond the median onset time of 7.2 months for adverse radiation effects after initial SRS.²⁶ This provides some assurance about the short-term safety of our approach. We hypothesized that the subsequent surgical removal of the irradiated target, including the margin of the surrounding brain tissue, might have contributed to this favorable toxicity profile. This approach is analogous to the standard preoperative SRS technique for newly diagnosed brain metastases;^6–19 however, a longer follow-up is needed to fully assess the long-term safety profile.

A crucial advantage of our strategy is the rapid implementation of Gamma Knife SRS at our institution. As neurosurgeons operate the system directly, SRS can be performed with short notice, even the day before urgent surgery. This flexibility allows the incorporation of SRS into oncological emergency protocols, which benefits patients with rapidly progressing symptoms or significant mass effects, whether for initial or repeat irradiation. Our approach also utilized single-fraction SRS, which reflects our institutional strategy for managing brain metastases. Patients who do not require immediate surgical intervention are typically managed with radiation therapy alone,^27–35 and preoperative SRS is reserved for cases that necessitate prompt tumor removal. In our series, we performed single-fraction SRS for preoperative treatment due to its advantage of rapid implementation. While the literature also reports the efficacy of FSRT in select patients,^13,18,19,36 the choice between single-fraction SRS and FSRT should be based on individual patient factors, tumor characteristics, the urgency of intervention, and institutional expertise.

Our case series reveals the feasibility and potential benefits of preoperative re-irradiation for recurrent brain metastases after SRS. This approach showed promising local control while maintaining an acceptable safety profile, with no cases of leptomeningeal spread observed during the follow-up period. The rapid implementation capability of SRS at our institution, performed by neurosurgeons, enables timely treatment for cases requiring prompt intervention, thus providing a crucial advantage in managing urgent cases. This study has several limitations that introduce potential biases, including its small sample size, retrospective nature, and relatively short follow-up period. Although these findings are promising, they should be interpreted with caution. Future research should focus on prospective studies with larger cohorts and longer follow-up periods to validate our findings, assess long-term outcomes and late toxicities, and compare preoperative re-irradiation with other management strategies for post-SRS recurrence.

Lessons

Preoperative re-irradiation, followed by resection, can be a valuable strategy for managing symptomatic post-SRS recurrent brain metastases. Our approach presents feasibility and safety without significant acute toxicity while offering promising local control and rapid symptom improvement. This strategy’s flexibility in managing cases requiring timely intervention potentially expands the treatment options for patients who might otherwise have limited choices in time-sensitive situations.

Acknowledgments

This work was supported by the Ministry of Education, Culture, Sports, Science and Technology (22K09223 and 23K15677 to M.O. and S.K.).

We thank Editage for English-language editing.

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

Conception and design: Kato. Acquisition of data: Kato, Hasegawa, Kuwabara, Naito, Mizuno, Sakai. Analysis and interpretation of data: Kato, Hasegawa. Drafting the article: Kato, Ohno, Kuramitsu. Critically revising the article: Kato, Ohno, Kuramitsu. Reviewed submitted version of manuscript: Kato, Hasegawa, Ohno, Kuramitsu. Approved the final version of the manuscript on behalf of all authors: Kato. Statistical analysis: Kato. Study supervision: Oishi. Pathological diagnosis: Kuwabara.

Correspondence

Takenori Kato: Komaki City Hospital, Komaki, Aichi, Japan. neuro.kato@gmail.com.