Reassessing the Management of Treatment-Induced Brain Necrosis in Metastatic Patients: Advancing Surgical Alternatives to Steroid Therapy

Abstract

The emergence of modern adjuvant therapies has significantly improved outcomes for patients with brain metastases. However, treatment-related side effects present an ongoing challenge, particularly, treatment-induced necrosis characterized by perilesional edema and inflammation. Standard management with steroids compromises the efficacy of otherwise efficacious immunotherapeutic approaches. This position paper critiques traditional management strategies that rely heavily on systemic corticosteroid therapy—often ineffective in providing lasting relief and associated with serious side effects—and proposes a paradigm shift that prioritizes surgical resection. Resection facilitates prompt edema reduction with a low recurrence rate of symptoms and mitigates the adverse effects of prolonged corticosteroid use. We propose increased consideration for resecting symptomatic radiation necrosis to facilitate improved efficacy of immunotherapies in patients with brain metastases.

ABBREVIATIONS:

cGAS

cyclic GMP-AMP synthase

LITT

laser interstitial thermal therapy

STING

stimulator of interferon genes

TME

tumor microenvironment.

Modern advancements in cancer-directed adjuvant therapies have provided great promise for patients with brain metastases. However, treatment effects after chemotherapy, immunotherapy, and/or radiation treatment remain a major challenge that can affect patient morbidity and mortality.¹ Treatment-induced necrosis, characterized by perilesional edema and inflammation, can create a space-occupying lesion that is difficult to distinguish radiographically from residual/recurrent tumor, particularly at the initial stages of posttreatment growth, and can lead to similar mass effect symptomology. Treatment necrosis also presents a significant challenge to the advancement of clinical trials.^2,3 For example, it may be perceived as progression, making patients ineligible for certain clinical trials. Including patients with treatment necrosis rather than true tumor progression or with recurrence complicates the accurate assessment of an emerging drug's antitumor activity, further complicating clinical trials.⁴ Many immunotherapy treatments^2,5,6 increase the incidence and prevalence of symptomatic treatment effect/edema and require tailored attention, such as in new response assessment guidelines after immunotherapy. At present, a standard for surgical intervention of symptomatic radiation necrosis in brain metastases has not been established.

The purpose of this position paper is to address the increasing concern of treatment-related necrosis in neuro-oncology, specifically in patients with brain metastases. We discuss how this condition has traditionally been managed by neuro-oncologists, radiation oncologists, and neurosurgeons, and present a perspective on the use of surgical intervention in the context of emerging treatments. Finally, we propose a path forward for surgical management of treatment necrosis in brain metastases.

DEFINING THE PROBLEM OF TREATMENT-RELATED NECROSIS

Treatment Effects

The primary management of brain metastases has predominantly included a combination of surgical resection, whole-brain radiation treatment, and/or radiosurgery.⁷ The benefit of systemic therapies in patients with brain metastases is poorly understood, and these are not routinely used.^8,9 However, advances in immunotherapeutic agents have shown promise—particularly in patients with brain metastases secondary to immunogenic tumors—and these agents now form the backbone of upfront systemic therapy in some cases.^10-12 Patients are followed serially with MRI to assess for recurrence. However, a significant problem in neuro-oncology is the appearance of new and usually progressive radiological changes after radiotherapeutic treatment without the presence of a significant tumor burden (only confirmed after resection), often broadly referred to as “treatment effect.” Treatment effect includes both treatment necrosis (involving tissue damage) and pseudoprogression (an inflammatory response), which refers to new or increased areas of contrast enhancement without significant tumor growth. Pseudoprogression may appear weeks to months after radiation treatment, while treatment necrosis appears months to years after treatment, most commonly following radiation. Although these 2 entities may differ based on their underlying pathophysiology, they are both difficult to differentiate from recurrent tumors on standard MRI, as they can present as heterogeneously enhancing lesions on T1-weighted imaging with surrounding T2/fluid-attenuated inversion-recovery signal.¹³ Pseudoprogression may occur concurrently with tumor progression, resulting in a mixed response, further complicating response assessment. Biopsy and/or resection remain the gold standard for distinguishing between tumor recurrence and treatment effects. However, biopsy often faces challenges in obtaining an adequate sample of viable tumor cells, particularly in tissues exhibiting varying degrees of treatment-induced changes and tumor burden.¹⁴

Treatment-Related Necrosis

The incidence of radiation necrosis in treated lesions has been reported to range from 3% to 30%.¹⁵ It is a relatively frequent consequence of fractionated stereotactic radiotherapy or single-session stereotactic radiosurgery treatment in brain metastases.¹⁶ It is also referred to as “treatment necrosis” to reflect its complicated nature, which extends beyond just radiation treatment as a cause. Several factors have been identified that may influence the development/extent of treatment necrosis: age, vascular comorbidities,¹⁷ and other therapeutic agents¹⁸ can all influence the presence and degree of treatment necrosis after or during radiation.¹⁹ For the purposes of this position paper, we will refer to treatment-related necrosis, whether induced by radiation or other treatment effects, as “treatment necrosis.”

There are only limited established guidelines for the histological diagnosis of treatment necrosis, and they commonly also vary between institutions,^20,21 but generally include the presence of varying degrees of vascular and endothelial reactivity with fibrinoid necrosis and vascular hyalinization, patches of perivascular lymphocytes, edema, and/or necrosis.²² Importantly, it is believed to be a sporadic event such that these histological features can progress in different locations within an irradiated field.²³ Progression is driven by proinflammatory cytokines within a macrophage-rich lesion,²⁴ such as through TNF-α²⁵ and NF-κB downstream signaling,²⁰ and especially the cyclic GMP-AMP synthase (cGAS) and stimulator of interferon genes (STING) pathway (cGAS-STING).^26-28

Given that treatment necrosis can occur sporadically anywhere within and/or around irradiated tissue, or may occur as enlargement of the initial tumor,^29-31 the clinical presentation is often heterogeneous.³² Although most patients are asymptomatic, up to 50% of patients¹⁸ present with symptomatic radiation necrosis with significant morbidity, including cognitive decline, increased intracranial pressure, seizures, and/or focal neurological symptoms secondary to mass effect.^19,32 The severity of these symptoms, combined with their unpredictable clinical course, presents a distinct challenge for oncological management.

TRADITIONAL MANAGEMENT IN THE ERA OF IMMUNOTHERAPY

Steroid Treatment and Classical Consequences

Few countries have standardized protocols for treatment necrosis.³³ Most commonly, a dexamethasone taper is prescribed.³⁴ However, if symptoms persist, then repeated rounds of steroids may be attempted.³⁴ Persistent steroid use leads to well-known Cushingoid symptoms, including myopathy and dysregulated blood glucose, cognitive dysfunction, bone demineralization, and gastric ulcers.³⁵ Furthermore, intravenous steroids can lead to severe liver damage due to direct cytotoxicity to hepatocytes in a likely dose-dependent manner.³⁶ Adverse events can be expected in up to half of the patients in the long term, regardless of dose,³⁷ and even when patients respond, a rebound in inflammatory effects may occur.³⁸ Although steroids may reduce the severity of treatment-induced necrosis symptoms,^34,39 the potential diminished immunological response is also counter-productive in the setting of immunotherapies.

Dexamethasone as a Barrier to Immunological Success

Systemic steroids are detrimental to an effective adaptive or innate immune response. This is because of several potential mechanisms, including a cytotoxic effect of steroids on T cells,^40-42 limiting their proliferation and differentiation in the tumor microenvironment (TME), as well as a fundamental alteration of myeloid cell states to an immunosuppressive phenotype.⁴³ Although the exact effects of steroids on immune function remain uncertain and often debated, their downstream impacts are extensive and, due to their pleiotropic nature, likely entail a range of significant consequences.

Metastatic brain tumor patients who are actively on immunotherapy or have previously received immunotherapeutic treatment are believed to be more likely to demonstrate symptomatic treatment necrosis.^5,6,26 This has been most notably demonstrated in patients with brain metastases treated with immune checkpoint blockade (ie, ipilimumab, pembrolizumab, nivolumab), particularly with dual-agent immunotherapy in patients with melanoma. The mechanism remains to be elucidated, but it may in part be due to the increased survival and surveillance of brain tumor patients on immunotherapy, leading to more frequent identification.⁴⁴ However, 1 possible etiology is the induction of the cGAS-STING pathway from double-stranded DNA breaks.⁴⁵ Induction of cGAS-STING in response to ionizing radiation results in several appropriate immune responses, including type I interferon signaling to bridge innate and adaptive antitumor immune responses to sites of DNA double-stranded breaks. However, cGAS-STING can become chronically overactivated⁴⁶ and may severely worsen treatment necrosis due to the induction of similar proinflammatory cytokines, which potentiate necrosis progression.^20,26

Dexamethasone remains a cornerstone in managing symptomatic treatment necrosis in patients with brain metastases, yet it can potentially restrict the efficacy of immunotherapy. Consequently, there is an urgent need to develop alternative management strategies.

Alternative Medical Strategies

Bevacizumab, a monoclonal antibody that targets and inhibits vascular endothelial growth factor, is 1 alternative to steroids.⁴⁷ By blocking vascular endothelial growth factor, bevacizumab aims to reduce angiogenesis and subsequent edema associated with treatment necrosis. Numerous studies have reported positive outcomes related to symptom relief and improvements in neurological function secondary to bevacizumab treatment.^48,49 Bevacizumab use has been complicated by extensive costs and a severe side effect profile, including gastrointestinal perforations, hypertension and, notably, hemorrhage. Furthermore, some patients with symptomatic treatment necrosis treated with bevacizumab have experienced worsening neurological function,⁵⁰ have become resistant to the drug,^19,51 and/or have a limited response.³⁹ Finally, the use of bevacizumab may temporarily preclude surgery and make MRI interpretation more difficult.

Other potential treatments have also been explored and reviewed extensively.¹⁹ This includes experimental treatments such as hyperbaric oxygen therapy,^52,53 various anti-inflammatory agents (ie, nonsteroidal anti-inflammatory drugs),⁵⁴ antiplatelet and anticoagulant agents with vitamin E,^55-57 and Boswellia serrata.⁵⁸ All proposed alternative agents remain experimental and have challenges that may limit their utility at the current time.

One treatment that has grown in interest as a minimally invasive surgical alternative to open resection is laser interstitial thermal therapy (LITT). LITT has been used successfully in the treatment of intra-axial brain tumors,⁵⁹ epilepsy,⁶⁰ and treatment necrosis.⁶¹ Its primary mechanism is through thermal ablation of the lesion and subsequent induction of coagulative necrosis, which varies by the degree of ablation delivered.⁶² Initially, LITT was used for deep-seated, eloquent lesions or those that have failed radiosurgery,⁵⁹ although its indications have rapidly expanded to now include cases of treatment necrosis that fail steroid treatment and progress in deep and superficial locations, regardless of tumor or treatment necrosis pathology.^61,63 LITT can increase inflammatory cytokines in the TME, such as the expression of interferon-γ, which can interfere with certain immunotherapeutic agents, such as viral replication and immune clearance of oncolytic viruses.⁶⁴ A thorough understanding of LITT's role in addressing treatment necrosis, along with the timing of immune landscape remodeling, is critical to optimizing immunotherapy strategies both before and after LITT interventions.

UPFRONT SURGERY FOR TREATMENT-INDUCED NECROSIS

Surgical resection offers a definitive treatment option for treatment necrosis in brain metastases,^65-69 but is typically reserved for cases where medical therapy has failed or when the inflammatory lesion is causing life-threatening mass effects, often in combination with steroids. In the modern neurosurgical era, we posit that surgical resection should be considered as the primary treatment option for treatment-induced necrotic lesions.

Surgical resection of treatment necrosis reduces edema with an exceptionally low rate of recurrence.^22,39,70,71 Patients initially treated with frontline steroids often experience treatment failure, a condition known as ‘steroid-refractory radiation necrosis,^39,72 and eventually require surgical resection,⁷¹ but only after enduring the adverse effects of prolonged steroid therapy. As the necrotic lesion continues to release a storm of inflammatory cytokines, delaying time to treatment only adds further chronic exposure of these immunosuppressive signals to the TME and further increases the size of the tumor, requiring additional immunosuppressive steroid treatment.^73-75 This may overwhelm the immune response and in theory, further reduce the targeted response against the tumor if not resected early on. In addition, in situations of pseudoprogression with a mixed response, surgical resection accomplishes both inflammation reduction and cytoreduction of any residual tumor.

Although upfront surgical resection is highly underused and the current time, there are many potential consequences that must be considered given that less invasive methods of treatment exist as discussed above. Most notably, a counterpoint to early surgical intervention for treatment necrosis is the potential for postoperative morbidity, and that the treatment necrosis may self-resolve. In addition, many patients at the time of treatment necrosis presentation may not be fit for surgery. However, modern neurosurgical techniques, minimally invasive surgeries,⁷⁶ and improved imaging and guidance have made these inflammatory lesions more amenable to earlier surgical resection with fewer risks.⁶⁶ Surgical resection offers the potential for significant steroid reduction, enabling some patients to decrease their steroid use by up to two-thirds,^22,65 thereby mitigating the side effects associated with prolonged steroid therapy. Decreased steroid use may enhance the effectiveness of immunotherapies in select cases. Importantly, patients may experience faster symptom relief from the debulking of symptomatic treatment necrosis. Another key advantage of surgical debulking is that it yields vital tissue to inform subsequent treatment decisions and to help define the overall effectiveness of experimental therapeutics where response assessment is challenging with contemporary diagnostic imaging. An important point to consider is the threshold for resection of the necrotic lesion. Many patients with treatment necrosis can have a relatively small, asymptomatic necrotic lesions that often resolve without treatment.^36,77 Therefore, we suggest upfront surgical resection for those lesions that are symptomatic, may become symptomatic, and/or are classically treated with prolonged steroid therapy.

ILLUSTRATIVE CASES

Case 1

A 61-year-old woman with a 1-year history of lung adenocarcinoma and multiple brain metastases, treated with stereotactic radiosurgery and pembrolizumab, developed progressive headaches and was found to have right frontal edema consistent with treatment effect. Steroids were required and pembrolizumab was held before surgery for pancreatitis. Pembrolizumab was temporarily discontinued, and the patient underwent surgical resection of the right frontal lesion. Pathology confirmed the absence of viable tumor, indicating treatment effect. After a multidisciplinary discussion including the patient, the patient was successfully weaned off steroids and restarted on pembrolizumab. Given the absence of viable tumor on pathology, postoperative surveillance was prioritized (Figure A-C).⁷⁸

FIGURE 1.

Examples of surgical management of treatment effect after radiotherapy and immune checkpoint inhibition in the setting of brain metastasis. Cases are described in the text above. A-C, Case 1 received stereotactic radiosurgery and pembrolizumab, whereas D-F case 2 received stereotactic radiation therapy, radiosurgery, and ipilimumab. A, Axial T2 FLAIR; B, preoperative axial T1 post contrast; C, postoperative day 1 T1 post contrast; D, axial T2 FLAIR; E, preoperative axial T1 post contrast; F, postoperative day 1 T1 post contrast. FLAIR, fluid-attenuated inversion-recovery.

Case 2

The second case involved an 88-year-old man with metastatic melanoma and multiple brain metastases, treated with stereotactic radiation therapy, radiosurgery, and ipilimumab. Four months after initiating ipilimumab, the patient developed a right visual field defect and impaired depth perception. The patient underwent surgery, and pathology revealed treatment effect and necrosis. After surgery, the patient was tapered off steroids and resumed checkpoint inhibitor therapy (Figure D and E). The study protocol was approved by the local Institutional Review Board, consent for the procedure was obtained, consent to publish cases was not required because it was covered by the retrospective Institutional Review Board protocol.

There are several treatment options that are available to brain metastatic patients with radiation necrosis, each of which has its own advantages and limitations. We summarize these treatments below (Table). This paper highlights that surgical resection is significantly underused and offers many important benefits, although it is crucial to recognize that less invasive therapies also play a key role in specific clinical scenarios. Corticosteroids, for example, have remained largely effective for immediate symptom relief in standard practice, though they come with well-established side effects, especially when used long-term such as Cushingoid features. Bevacizumab, though limited by its cost and potential side effects such as hemorrhage, can reduce symptomatic edema in patients with minimal mass effect symptomology. This approach remains valuable in cases where surgical intervention is not feasible, such as deep-seated lesions and patients unfit for surgery, or when lesion progression is less aggressive. However, when patients are candidates for surgery, we advocate for earlier consideration of surgical resection, because it provides definitive symptom relief, adequate tissue sampling, minimizes steroid use, and allows for better integration with immunotherapy. A tailored, proactive approach to treatment—prioritizing surgery when appropriate—can significantly improve patient outcomes in the management of radiation necrosis.

TABLE.

Weighing the Advantages and Disadvantages of Common Treatments for Radiation Necrosis in Brain Metastatic Patients

Treatment modality	Pros	Cons	Considerations
Steroids (eg, dexamethasone)	Immediate symptom relief (eg, reduced edema)	Long-term side effects (eg, Cushingoid symptoms)	Short-term use for symptom management
	Widely available and easy to administer across centers	Immunosuppressive effects, reducing efficacy of immunotherapy	Long-term treatment limited by adverse effects
Bevacizumab	Effective for reducing edema while avoiding Cushingoid features	Expensive and limited access	Best used for lesions with minimal mass effect
	Can improve neurological function in some patients	Severe side effects (eg, hypertension, hemorrhage)	Potential delay of surgery
		Can interfere with surgery (requires washout period)
Hyperbaric oxygen treatment	Restores oxygen concentration in the necrotic to increase blood flow	Expensive with limited accessibility	May be considered when other treatments are ineffective
	Noninvasive	Evidence limited to case reports with no long-term data	Not widely available in all centers
LITT	Minimally invasive—effective for deep-seated lesions	Can increase inflammatory cytokines, which may interfere with immunotherapy	Best for recurrent lesions after radiosurgery
	Potentially less morbidity compared with resection	Limited evidence for long-term outcomes	Requires careful management of timing with immunotherapy
Surgical resection	Definitive treatment with fast symptom relief	Invasive with potential postoperative morbidity	Recommended for symptomatic lesions or when other treatments fail
	Immediately reduces edema and inflammation	Standard risk of complications from surgery (infection, bleeding)	Should be considered earlier to avoid prolonged steroid use and immunosuppressive effects
	Tissue diagnosis

LITT, laser interstitial thermal therapy.

CONCLUSION

The management of symptomatic treatment necrosis in brain metastases in the era of immunotherapy and immunomodulation requires a paradigm shift away from traditional steroid-based therapy. High-dose systemic steroid therapy, typically involving dexamethasone, poses significant challenges, including long-term toxicity and the potential to compromise the efficacy of immunotherapies. Surgical resection of brain metastases before adjuvant radiation significantly reduces the development of symptomatic radiation necrosis compared with upfront radiation to the nodular disease. Moreover, resection of either recurrent tumor or treatment effect provides immediate relief from mass effect and removes the nidus triggering edema formation. In addition, surgery provides invaluable tissue to inform treatment decisions and clinical trial efficacy. Modern neurosurgical techniques enable safer and often minimally invasive resection of these lesions. We advocate for a more proactive consideration of surgical resection in the upfront treatment of symptomatic treatment-induced necrosis when surgery is safely feasible, especially as promising immunotherapies continue to emerge. Controlled trials to support this clinical observation may help elucidate the impact of surgery in this setting.

Funding

This study did not receive any funding or financial support.

Disclosures

Joshua D. Bernstock holds equity in Treovir, Inc and UpFront Diagnostics. He is a co-founder of Centile Bioscience, serves on the scientific advisory boards of NeuroX1 and QV Bioelectronics, and is a member of the board at Synaptive Medical. Lennard Spanehl receives financial support from the German Research Foundation (DFG). Ayal Aizer is a consultant for Seagen and Novartis and receives financial support from Varian and NHT. Gregory K. Friedman receives financial support from Pfizer, Eisai, and E.R. Squibb & Sons and has equity interest in Synaptive Medical, Inc. The other authors have no personal, financial, or institutional interest in any of the drugs, materials, or devices described in this article.