Neuro-Oncology
Overview of neuropathology updates for infiltrating gliomas
Oct. 11, 2024
MedLink®, LLC
3525 Del Mar Heights Rd, Ste 304
San Diego, CA 92130-2122
Toll Free (U.S. + Canada): 800-452-2400
US Number: +1-619-640-4660
Support: service@medlink.com
Editor: editor@medlink.com
ISSN: 2831-9125
Toll Free (U.S. + Canada): 800-452-2400
US Number: +1-619-640-4660
Support: service@medlink.com
Editor: editor@medlink.com
ISSN: 2831-9125
Worddefinition
At vero eos et accusamus et iusto odio dignissimos ducimus qui blanditiis praesentium voluptatum deleniti atque corrupti quos dolores et quas.
Radiation therapy is an effective therapy for many malignancies and benign conditions. However, radiation therapy can potentially cause early and late toxicities in the central nervous system. These include radiation necrosis, cerebrovascular disease, cognitive deficits, endocrinopathies, encephalopathy, myelopathy, plexopathy, radiculopathy, neuropathy, and secondary tumors. This article discusses in detail radiation complications in the central nervous system, prevention of radiation toxicity, and therapeutic options.
With the widespread adoption of conformal intensity modulated radiation therapy (IMRT), volumetric modulated arc therapy (VMAT), particle therapy, and stereotactic radiation approaches, it has become more possible to limit and prevent radiation toxicity. In recent years, novel methods to prevent radiation toxicity have been reported. NRG Oncology CC001 showed that hippocampal-sparing whole brain radiation therapy decreases neurocognitive injury compared to conventional whole brain radiation therapy. Prospective studies are now investigating the role of sparing other important structures involved in neurocognition, such as the fornix, corpus callosum, and amygdala. Spine stereotactic body radiation therapy (SBRT) has become more prevalent in the treatment of spinal osseous metastases and has been demonstrated to be generally safe, with no difference in toxicity including in rates of myelopathy, when compared with conventionally fractionated radiation therapy in recently published prospective trials.
The diagnosis and treatment of radiation necrosis continues to improve, with more imaging techniques and treatment options available. Advancements in diagnosis include the application of perfusion magnetic resonance imaging (MRI) to determine relative cerebral blood volume (rCBV) and newer imaging modalities like amino acid positron emission tomography (PET). Contemporary treatment options for radiation necrosis include bevacizumab and laser-interstitial thermal therapy (LITT).
Treatment of late cognitive impairment due to central nervous system radiation continues to be an elusive goal.
• When radiation therapy is used to treat primary or metastatic central nervous system (CNS) diseases, or non-CNS targets located close to neural structures, side effects to the normal neural tissues can occur. | |
• When practicing within accepted constraints, the acute and subacute complications of radiation therapy are generally mild, transient, or treatable with corticosteroids. | |
• In contrast, the late complications of radiation therapy are generally progressive and often permanent. | |
• The incidence and severity of radiation-induced CNS complications varies with the radiation dose, fractionation scheme, volume of tissue irradiated, and target location; patient age; underlying diseases (malignant and nonmalignant); concomitant treatments; comorbidities; and length of survival after completion of radiation treatment. | |
• In general, the risks of radiation-related CNS side effects are balanced against the risk of progressive or recurrent disease. |
Older radiation therapy techniques used to treat primary or metastatic nervous system diseases, or structures adjacent to neural structures, caused damage to the nervous system. The most dramatic example of this type of injury, brain radiation necrosis, was first recognized in 1930, soon after radiation was first used therapeutically for brain tumors (23). Since that time, a spectrum of injuries throughout the central and peripheral nervous system has been identified, and some of the details of specific syndromes have been elucidated. Despite this heightened awareness, the neurologic complications of radiation therapy continue to occur because individual tolerances to radiation are variable, safe radiation thresholds are not precisely known, latency to development of injury range between days to years, and risks are altered by use of chemotherapy, other systemic therapies, or preexisting disease. Conventional and stereotactic radiation treatment cause similar toxicities, though dose per fraction, volume irradiation, location, and other dosimetric factors may affect severity and timing of injury. Despite increasing understanding and ability to prevent CNS radiation toxicity, the incidence of radiation-related nervous system side effects are likely to increase as patients survive longer.
Radiation-related nervous system side effects can affect every level of the nervous system and can be classified either anatomically or temporally (Tables 1 and 2).
Brain |
• Acute encephalopathy |
Neuroendocrine |
• Hypothalamic, pituitary, and thyroid hypofunction |
Cranial neuropathies (including optic neuropathy) | |
Spinal cord |
• Subacute (transient) myelopathy |
Brachial and lumbosacral plexuses |
• Transient brachial plexopathy |
Peripheral nerves |
• Perineural fibrosis |
Cerebral vascular disease |
• Intracranial arterial occlusive disease |
Acute (early) |
• Acute encephalopathy |
Subacute |
• Subacute ("early delayed") encephalopathy |
Late (chronic) |
• Delayed cerebral radiation necrosis |
Accurate diagnosis based on clinical manifestations is critical in order to exclude other potentially treatable disorders, to prevent unnecessary diagnostic procedures and inappropriate antineoplastic therapy, and to allow for meaningful intervention.
Acute complications. Common acute side effects of radiation therapy include fatigue, headache, loss of appetite, nausea, and vomiting, and re-emergence or worsening of preexisting neurologic symptoms.
The most common and often most debilitating effect of cranial radiation therapy is progressive fatigue (usually between weeks 3 and 5 in fractionated radiotherapy), which may often persist for several weeks after the completion of therapy.
In the setting of therapeutic radiation therapy, acute radiation encephalopathy is a rare occurrence. It is characterized by lethargy, possibly accompanied by new or progressive focal deficits, headache, nausea, vomiting, fever, and seizures. Onset is within the first two weeks of treatment (may be as early as within 24 hours).
Subacute complications. Lethargy, impaired memory retrieval, and cognitive and behavioral changes may develop within one month from initiation of radiation therapy and extending to within 4 months following radiation therapy completion ("early delayed encephalopathy"). Ataxia, nystagmus, nausea, vomiting, and dysarthria may be present when the brainstem has been irradiated. Patients may experience transient hearing loss from soft tissue swelling and/or inflammatory otitis media.
Subacute myelopathy may occur following radiation therapy to the cervical (less often thoracic or thoracolumbar) spinal cord. Lhermitte sign, transient electric shock sensation with next flexion, may be the sole symptom and peaks at 4 to 6 months (range 1 to 30 months) after radiation exposure.
Late (chronic) complications. Late complications of CNS radiation therapy include radiation necrosis, cognitive deficits, endocrinopathies, cranial neuropathies, myelopathy, cerebrovascular disease, and secondary tumors.
Patients with delayed cerebral radiation necrosis may present with headache, personality change, focal deficits, and seizures. These symptoms typically develop insidiously 4 months to 4 years or more (median 14 months) after treatment. Occasionally, gait impairment, incontinence, and dysarthria can occur. Imaging findings consistent with radiation necrosis are discussed in the Diagnostic workup section below. Rarely, the presentation is fulminant, and even less commonly, established radiation necrosis may be complicated by acute hemorrhage (12).
A more diffuse late brain injury manifests clinically in progressive cognitive impairment, characterized by deficits in memory, attention, and executive function, fatigue, and even personality change and dementia. Age at irradiation is one primary determinant of this sequelae. The negative impact on neurocognition is more pronounced in children and older adults, particularly over the age of 70. In children, even low doses of cranial irradiation can be associated with declines in IQ and academic achievement. Children may also exhibit memory deficits, fine motor and visual-spatial dysfunction, and psychological disturbances (58; 54).
Radiation to the sellar, parasellar, or hypothalamic regions can result in endocrine dysfunction. Among adults, endocrine dysfunction can occur insidiously, developing over months to years and often undiagnosed. Children who receive brain radiation are even more vulnerable. Patients should undergo regular endocrine evaluation if their treatment was to this region, and they should receive prompt treatment for pituitary deficiencies. The most common radiation-induced endocrinopathies are hypothyroidism and growth hormone deficiency.
Radiation induced optic neuropathy can occur after irradiation of the optic apparatus, including the retina, optic nerve, chiasm and pituitary region, or optic radiation. Painless, progressive, monocular visual loss, or constriction of visual fields is the typical presentation. Altitudinal field cuts are common; "dimming" of vision or "spotty" visual loss are typical patient descriptions. The presence of pain or homonymous field defects weigh strongly against the diagnosis. Onset ranges from three months to several years (median 11 months).
Although the optic nerves are the most sensitive of the cranial nerves to radiation therapy, other cranial neuropathies can develop following exposure to therapeutic radiation (04). Radiation-induced cranial neuropathy occurs 1 to decades (mean 5.5 years) following radiation therapy (generally for head and neck or orbital tumors). In order of frequency, cranial nerves XII, XI, X, V, and VI are affected.
Chronic progressive myelopathy is the delayed spinal cord syndrome corresponding to cerebral radiation necrosis (27; 78). It most commonly occurs following radiation therapy of tumors in the chest, mediastinum, cervical region, or head and neck. The syndrome frequently presents with ascending paresthesias, dysesthesias, or sensory loss in one or both lower extremities, followed by weakness and signs of myelopathy. A partial transverse myelitis or Brown-Sequard syndrome is possible, as is disturbance of sphincter function. Symptoms begin 3 to 30 months or more (median 20 months) after radiation treatment, and progression is usually gradual over weeks to months.
The cerebral vasculature can be damaged by radiation. Depending on the portion of the vascular tree affected and the type of vascular lesion, transient ischemia, strokes, or hemorrhage can occur. Onset may be 10 or more years following radiation exposure, necessitating long-term surveillance (76). Children are more vulnerable than adults in their risk of developing radiation-induced vasculopathy. Radiation-induced cerebral microbleeds and large vessel cerebral vasculopathy have been reported in pediatric patients undergoing radiation therapy for brain tumors (59; 38). Radiation-induced cavernous malformations are another manifestation of radiation-related cerebrovascular disease, associated with brain doses in the range of 45 to 60 Gy (61). Moyamoya syndrome is a rare complication of cranial irradiation, particularly in children who received higher doses to the circle of Willis (84). Stroke-like migraine attacks after radiation therapy (SMART) syndrome is a rare late complication, associated with radiotherapy greater than or equal to 50 Gy, and characterized by migraine-like headaches, stroke-like deficits, and seizures. MRI findings include cortical gyriform enhancement with corresponding T2-FLAIR hyperintensity and diffusion restriction inconsistent with vascular territories (07; 88). Clinical symptoms and radiologic signs are reversible though incomplete recovery or recurrent episodes may occur (63). Radiation that is administered to the neck can result in delayed carotid atherosclerosis. Tumor embolization to the brain is a rare cause of stroke.
Radiation-induced CNS tumors constitute a small but serious risk for patients undergoing radiotherapy for the management of cerebral neoplasms. Studies of pediatric survivors who developed second brain tumors have found that meningiomas are the most common. The average latency period for the appearance of the second tumor was eight years, but meningiomas had a longer latency period, ranging from 16 to 30 years in one study. Other common second cancer histologies include peripheral nerve sheath tumors and gliomas. The symptoms of radiation-induced intracranial tumors are indistinguishable from those of their nonradiation-induced counterparts, but radiation-induced intracranial tumors are typically more aggressive pathologically and clinically. Radiation-induced meningiomas, for example, are higher grade with high labeling indices, may appear with multiple synchronous tumors, and tend to recur more frequently and earlier after gross total resection (67; 82). NF2 gene rearrangements, which appear to be specific to radiation-induced meningiomas, have been described in approximately half of patients with such tumors (01).
In addition to CNS-specific side effects, a constellation of acute or late nonneurologic symptoms are common after cranial irradiation (Table 3).
• Hair loss |
The acute and subacute complications of radiation are generally mild, transient, or treatable with corticosteroids. The symptoms of early delayed encephalopathy begin to resolve within 2 to 4.5 months of onset (03), whereas those of subacute myelopathy disappear within four months of onset. Recognition of these acute and subacute syndromes permits early intervention or reassurance and may obviate the need for invasive or expensive diagnostic interventions and therapy directed at presumed tumor recurrence.
In contrast, the late complications of radiation are generally progressive and severe. They typically result in significant disability and are of particular concern in patients with potentially curable disease (eg, childhood acute leukemias, intracranial germ cell tumors, pituitary tumors, and meningiomas), or tumors compatible with long survival (eg, oligodendrogliomas and limited brain metastases with well-controlled systemic disease).
Although the histopathology of some forms of radiation-induced nervous system injury has been described, the etiologies of many of the conditions described above are still not well understood.
Acute toxicity (occurring within hours to weeks of treatment) is related to edema and inflammation. This theory is supported by neuroimaging data and the clinical observation of steroid responsiveness. Newer data suggest that inflammation may be associated with early injury to neuronal lineages, accessory cells and their progenitors in the brain, and supporting structures. Late injuries (occurring months to decades after treatment) has multifactorial causes, including vascular injury, stem cell depletion, and changes in the neural microenvironment (45; 40).
Cerebral radiation necrosis and radiation myelopathy are the best characterized radiation-related syndromes from a histopathologic standpoint. Fibrinoid necrosis, endothelial proliferation, hyalinization and thickening of vessel walls, adventitial fibroblast proliferation, thrombosis, telangiectasias, and multinucleated astrocytes all contribute to coagulative white matter necrosis.
Reversible radiation myelopathy is characterized by prominent demyelination with axonal loss but only minimal vascular injury (41).
The histologic alterations in diffuse late brain injury with cognitive decline are less dramatic and less specific. Gliosis, neuronal loss, and white matter spongiform changes predominate (19). Vasculopathy in the form of endothelial damage, microvascular abnormalities, and disruption in cerebral blood flow are seen. Alterations in dendritic spines, which may disrupt synaptic neurotransmission, have also been observed. There is neural stem cell loss in the hippocampi in particular (45; 40). Imaging studies have found certain brain regions to demonstrate greater postradiation changes, resembling accelerated aging in some cases. Particularly radiosensitive brain substructures may include the hippocampi, fornix, and corpus callosum (33; 69).
Radiation-induced optic neuropathy is similarly thought to be due to damage to endothelium and neuroglial cell progenitors leading to vaso-occlusion, demyelination, and neuronal degeneration (42).
The incidence of radiation-induced nervous system complications varies with the radiation dose, volume of tissue irradiated, fractionation scheme, and radiation technique; degree of edema; patient age; underlying diseases (malignant and nonmalignant); concomitant treatments; and length of survival after completion of radiation treatment. Risk of deficits after cranial irradiation increases with higher radiation therapy dose, larger fraction size, larger field size, and extremes of age at time of treatment. As these same factors are also associated with higher burden of disease, the causative relationship is confounded. Analyses of neurocognitive function are also confounded by factors such as surgery, chemotherapy (particularly concurrent and sometimes even temporally offset radiosensitizing agents), immunotherapy, targeted agents, and tumor characteristics. Finally, patients’ underlying conditions such as hypertension, diabetes, preexisting neurologic disease, and nononcologic medications may affect prevalence and timing of CNS radiation toxicity.
Radiation-related white matter changes are common on CT and MRI scans (16), and as many as 20% of patients with the radiographic correlates of diffuse late brain injury develop frank radiation-induced dementia (18).
The dose-response relationship for brain toxicity is not straightforward due to the confounding factors discussed above and the challenges of evaluating complex dose distributions (39). As a rule, incidence increases and latency decreases with higher total doses, higher fraction size, and larger volumes of treated nervous system (47).
A significant determinant of cognitive injury is the volume and substructure of brain irradiated. Targeting the whole brain is more likely to cause cognitive injury and chronic fatigue, and is associated with greater severity of injury compared with partial brain irradiation. When whole brain radiation therapy is necessary, reducing dose to the hippocampi, which harbor neural stem cells and are thought to be play a key role in mediating radiation-induced cognitive toxicity, has been demonstrated to result in better cognitive preservation (10).
The incidence of radiation necrosis varies with dose per fraction and volume irradiated, among other factors. For stereotactic radiation, higher rates are reported with single-fraction stereotactic radiosurgery, around 20% with median of dose 20 Gy, compared with fractionated stereotactic radiotherapy, up to 8%. Limiting normal brain volume exposed to 12 Gy (V12Gy) is recommended to reduce the risk of necrosis with stereotactic radiosurgery (28). Quantec estimates a 5% risk of necrosis with a conventionally fractionated dose of 72 Gy (46). In the reirradiation setting, radiation-induced normal brain tissue necrosis is found to occur at normalized total dose of (cumulative) greater than 100 Gy (49).
Immune checkpoint inhibitors, given either concurrently or sequentially in combination with radiation, may increase risk for radionecrosis. Retrospective studies report conflicting results. Some studies have found an increased risk (13; 48), reporting radionecrosis rates as high as 20% of patients, whereas others have not (22).
Radiation-induced optic neuropathy is a rare phenomenon. With hypofractionated radiotherapy, the Qualitative Analysis of Normal Tissue Effects (QUANTEC) recommends a single-fraction radiotherapy limit of 12 Gy in one fraction, 19.5 Gy in three fractions (83), and 25 Gy in five fractions (AAPM Task Group 101), with which there is less than 1% risk of neuropathy (31). Incidence increases with higher biologically effective dose. With conventional fractionation radiotherapy, doses below 54 Gy are considered relatively safe (51). In a large retrospective study of mixed proton and photon radiotherapy, cumulative incidence of optic neuropathy was 1% with doses less than 59 Gy and 5.8% in patients receiving 60 Gy and above. Higher point dose to the optic pathway, female sex, and older age were associated with increased risk of optic neuropathy (44).
Based on 3D dose data, the risk of brainstem injury markedly increases above 54 Gy to the entire brainstem, and 59 Gy to more than 10 ml of the brainstem (52).
Because the symptoms of neuroendocrine dysfunction may be subtle or difficult to distinguish from other tumor-related symptoms, it may be underdiagnosed. At least one biochemical endocrine abnormality has been detected in two thirds to three quarters of children and adults at some point following radiotherapy (15; 02). In a case-control study, 26% of patients had biochemical evidence of hypothalamic hypothyroidism, 32% showed evidence of hypothalamic hypogonadism, and 29% had hyperprolactinemia (02). Children are more susceptible to radiation-related neuroendocrine dysfunction than adults. Increased activity of the hypothalamic or pituitary axis, concurrent cisplatin or nitrosourea chemotherapy, and higher radiation therapy doses (especially above 18 Gy) also predispose to hypopituitarism.
The QUANTEC working group provides an estimate for myelopathy when applying radiation to the full-thickness cord in conventional fractionation (1.8 to 2 Gy). It reports the risk of myelopathy of 1% for 54 Gy and 10% for 61 Gy. A strong fractionation dependency was shown with an estimated α/β of 0.87 Gy. Most clinicians apply far stricter dose constraints in the range of 45 to 50 Gy because of the devastating consequences if myelopathy should occur.
Spine stereotactic body radiation therapy (SBRT) has been increasing in utilization because randomized clinical trials demonstrated improved pain control with SBRT over conventionally fractionated radiation therapy for some spine metastases (75; 73). Prospective studies have reported no significant difference in myelopathy, plexopathy, or vertebral compression fracture between patients treated with spine stereotactic body radiation therapy and those treated with conventionally fractionated radiation therapy (75; 73; 91). The following spinal cord point maximum doses are estimated to be associated with 1% to 5% risk of radiation myelopathy: 12.4 to 14 Gy in one fraction, 17 Gy in two fractions, 20.3 Gy in three fractions, 23.0 Gy in four fractions, and 25.3 Gy in five fractions (74).
The incidence of radiation-induced secondary tumors is directly related to the radiation dose and volume of tissue irradiated and inversely related to the age of radiation exposure. Excess relative risk (ERR) has been estimated at 0.2 to 6 per Gy for any secondary brain tumor, 0.6 to 5 per Gy for meningioma and 0.08 to 0.6 per Gy for glioma (09).
Preventive strategies have mostly been the result of improvements in radiation delivery technology. These include advancements in imaging defining targets for irradiation, and more conformal treatment planning and precise targeting with IMRT/VMAT, stereotactic radiosurgery (SRS)/stereotactic radiotherapy (SRT), and/or particle therapy. Greater knowledge of clinical radiation tolerances helps to guide future treatments. Additionally, there is application of aggressive control of increased intracranial pressure, modifications in radiation dose when concurrent chemotherapy is being used, and exploration of alternatives to irradiation in the very young and very old have led to reduced radiation associated toxicities.
Memantine has shown some promise in providing neuroprotection when given concurrently with radiotherapy. In one randomized clinical trial, memantine delayed time to cognitive decline and patients receiving memantine had improved executive function and processing speed, and a trend toward improved memory recall, compared to patients receiving placebo (11).
Hippocampal-avoidance whole brain radiotherapy (HA-WBRT), conformal avoidance of bilateral hippocampi during whole brain radiotherapy, has been shown to reduce the long-term cognitive toxicity of whole brain irradiation. A phase 2 clinical trial initially demonstrated less decline in memory recall and quality of life compared to historical controls (29). The phase 3 clinical trial NRG Oncology CC001 demonstrated that patients receiving HA-WBRT and memantine had significantly lower rates of and longer time to neurocognitive decline than patients receiving standard whole brain radiotherapy and memantine (10). Avoidance of additional brain structures involved in memory and cognition like the amygdala, corpus collosum, and fornix is a promising area of investigation for further preserving neurocognitive function (14; 66).
Retrospective studies have suggested that sparing brain tissue using proton therapy may also help to preserve cognitive function and other physiological processes (90; 30; 81). Ongoing clinical trials are investigating this possibility. There is emerging evidence from in vivo animal studies that FLASH radiation therapy (FLASH-RT), ie, treatment delivered in fractions of a second using high dose rates of greater than 100 Gy/second, may spare normal tissues without compromising tumor control. In studies of mice, high therapeutic doses could be delivered while minimizing neurocognitive toxicity with FLASH-RT (08; 80; 57). The mechanism of this superior normal tissue sparing has not been fully elucidated but may be related to a protective effect of radiochemical oxygen depletion, decreased generation of reactive oxygen species, and modification of the immune response to radiation (56; 87).
In the setting of increasing use of immunotherapy and possible increased radionecrosis with concurrent immunotherapy, ongoing studies are examining if reduced-dose stereotactic radiosurgery with immunotherapy provides comparable efficacy with decreased toxicity. Interim analysis of one such trial, on which patients received 18 Gy for lesions 0 to 2 cm, 14 Gy for lesions 2.1 to 3 cm, and 12 Gy for lesions 3.1 to 4 cm, has demonstrated excellent local control rates with no radiographic radionecrosis at six months (53).
Preoperative stereotactic radiosurgery as a strategy to maintain the local control provided by postoperative stereotactic radiosurgery while reducing radionecrosis and iatrogenic leptomeningeal spread is also under investigation. Retrospective studies report low rates of radionecrosis (5% to 7%) and leptomeningeal spread (3% to 7%) with lower dose preoperative stereotactic radiosurgery (64; 68). Phase III randomized clinical trials comparing pre- and postoperative stereotactic radiosurgery are underway.
In general, the most frequent and most pressing diagnosis competing with radiation-related nervous system injury is recurrent tumor. For specific syndromes, however, other neurologic and nonneurologic conditions may complicate the differential (Table 4).
Radiation-related sequelae |
Alternative etiologies |
Acute and subacute encephalopathy and myelopathy |
• Metabolic encephalopathy |
Radiation necrosis |
• Tumor recurrence |
Diffuse late brain injury |
• Tumor recurrence |
Neuroendocrine dysfunction |
• Psychiatric disorders |
Optic neuropathy |
• Drug effect |
Cranial neuropathy |
• Tumor recurrence |
Chronic progressive myelopathy |
• Epidural or intramedullary metastasis |
Motor neuronopathy |
• Effects of chemotherapy |
Plexopathy |
• Metastatic tumor |
Cerebrovascular disease |
• Nonbacterial thrombotic endocarditis |
When new symptoms develop over days or weeks, are mild, or improve over the weekend break from radiation therapy, a presumptive diagnosis of acute radiation toxicity and an empirical increase of steroid dose are reasonable. Marked or abrupt deterioration or fluctuating symptoms raise the possibilities of tumor progression, tumor-associated hemorrhage or increased edema, obstructive hydrocephalus, unrecognized seizures, or infection (intracranial or systemic).
The most common diagnostic quandary arises when a patient presents with a recurrent mass in the same location as the original tumor. When the new lesion develops later, greater than 6 months to years after the completion of radiation therapy, recurrent tumor is most likely, but the possibility of delayed radiation necrosis must be considered (even years later) particularly in the context of new systemic therapy with CNS activity, such as immunotherapy or targeted therapies. MRI or other imaging described in the next section may differentiate the process.
The term pseudoprogression refers to increased or new areas of enhancement of treated areas on MRI within months after radiation therapy or systemic treatments with CNS activity such as chemotherapy or immunotherapy, due to transient blood-brain barrier disruptions. Patients are often asymptomatic or minimally symptomatic despite imaging changes. The imaging changes usually shrink or resolve spontaneously within a few months. It is often difficult to distinguish pseudoprogression from radionecrosis and true progression, though imaging changes over time, radiation dose to the areas involved, and novel imaging techniques may help with diagnosis (20; 05).
Diffuse late brain injury can also present both a clinical and radiographic diagnostic dilemma. Clinically, leptomeningeal disease, encephalitis (infectious or paraneoplastic), concurrently administered drugs (anticonvulsants, steroids, and analgesics), metabolic abnormalities, systemic infection, and depression may be considered in the differential diagnosis. Chemotherapy alone, particularly high-dose chemotherapy, can produce cognitive, neurophysiological, and radiographic changes that mimic the effects of radiation (55; 77). One smaller, prospective trial (GLIO-CMV-01 study) suggested encephalopathy caused by cytomegalovirus viremia may be an overlooked and underdiagnosed condition (26). Rarely, radiographic confusion with periventricular small vessel disease, multiple sclerosis, progressive multifocal leukoencephalopathy, or transependymal CSF resorption in the setting of hydrocephalus can occur. Ideally, a diagnosis should not be made based on radiographic findings alone.
However, hemorrhages occurring in the brain parenchyma are more commonly caused by bleeding tumor or coagulopathies related to a patient’s malignancy or antineoplastic therapy.
Underlying patient conditions that predispose to radiation injury, such as lupus, connective tissue disease, or radiation sensitivity syndromes (ie, ataxia, telangiectasia, and NF1), may increase the risk of CNS radiation toxicity.
In general, a focused history and physical, review of medications particularly those with sedative or psychoactive properties, CNS imaging, and laboratory studies such as comprehensive metabolic panel and complete blood count, should be part of the initial diagnostic work up.
A CT or MRI scan is an appropriate first diagnostic step when symptoms suggest the possibility of radiation acute setting; CNS imaging can reveal edema or intracranial bleeding.
Radiation necrosis has the appearance of ring enhancing mass lesion on neuroimaging studies (35) and may be rapidly progressive. T1 and T2-weighted imaging, diffusion weighted imaging (DWI), and apparent diffusion coefficient (ADC) are recommended conventional MRI sequences for diagnosis. Highly cellular tumors may exhibit lower ADC due to relatively restricted motion of water, compared with radiation necrosis. With standard MRI neuroimaging, the radiographic appearance of radiation necrosis may be indistinguishable from primary or secondary brain tumors.
Efforts have been made to better identify radiation necrosis using advanced imaging techniques (05; 06; 50). Perfusion MRI with relative cerebral blood volume (rCBV) can show tumor angiogenesis and neovascularization, increased with tumor growth but decreased with radiation necrosis. MR spectroscopy using ratios of metabolites such as increased lipid/choline, increased lactate/creatine, or decreased choline/creatine has shown some promise in diagnosing radiation necrosis. Positron emission tomography (PET) using L-3,4-dihydroxy-6-[18F]-fluorophenylalanine (F-DOPA), [11C]-methyl-L-methionine (C-MET), O‐(2‐[18F]fluoroethyl)‐L‐tyrosine (18F‐FET), 18F-fluciclovine, and other radiotracers have been demonstrated to aid diagnosis. Generally, amino acid PET has been found to be superior to fluorodeoxyglucose (FDG) PET in sensitivity and specificity of diagnosing radiation necrosis. Single photon emission tomography (SPECT) using thallium-201 or technetium-99 may be another nuclear medicine modality that can aid diagnosis. Algorithms and guidelines are available that take into account multi-modality imaging results to help establish a diagnosis of radiation necrosis (05; 50). Radiomics-based prediction models, some employing artificial intelligence, have also been proposed to help improve diagnostic accuracy (65; 21; 62). Biopsy is ultimately the gold standard for diagnosis of radiation necrosis but is often reserved for persistent or severe symptoms despite nonsurgical treatments.
In the absence of a confounding neurologic disease, such as multiple sclerosis, or if tumor recurrence is not a concern, additional diagnostic investigations in cases of subacute myelopathy are unnecessary. Spinal cord atrophy or fusiform enlargement of irradiated segments of spinal cord are occasionally seen on MRI or myelography in patients with radiation myelopathy, but MRI reliably differentiates tumor involvement of the epidural from the intramedullary space and from leptomeningeal metastases, which are the chief competing diagnoses. The CSF protein may be elevated in all four disorders, but a pleocytosis, a hypoglycorrhachia, and a positive cytology distinguish leptomeningeal spread of tumor.
In patients with radiation-induced optic neuropathy, careful review of the radiation treatment plan, in conjunction with neuroradiographic imaging and neuroophthalmology evaluation, are important. A characteristic MRI appearance for radiation-induced optic neuropathy consists of discrete focal areas of enhancement along the intracranial optic nerve. Ophthalmic examination may uncover loss of visual acuity and color vision, visual field defects, and pallor of the optic discs on fundoscopy.
In the setting of possible radiation-induced cranial neuropathy, CT scanning may be diagnostic of tumor recurrence at the skull base, although prolonged and careful observation is frequently required. Because leptomeningeal disease, sarcoidosis, Lyme disease, basilar meningitis, and paraneoplastic encephalomyelitis can also present with cranial neuropathies, a lumbar puncture and brain MRI also may be required.
Furthermore, when there is radiation dose exposure to the pituitary/hypothalamic axis, it is advisable to routinely perform a baseline screening (T3U, T4, TSH, FSH, LH, prolactin, testosterone, and in children, GH) prior to radiation therapy, and screen routinely, typically annually, thereafter for radiation-induced neuroendocrine deficits.
As with nonradiation-associated cerebral vascular disease, CT or MRI can uncover infarcts, and conventional or MR angiography helps distinguish among potential etiologies. MR angiography may also help diagnose accelerated carotid atherosclerosis in irradiated segments of vasculature not otherwise commonly involved (eg, the proximal common carotid artery, internal carotid artery distal to the bifurcation, and small and medium-sized intracranial arteries) (60).
For patients with suspected late cognitive decline due to radiation therapy, neuropsychiatric and neurocognitive evaluation are useful tools for diagnosis and monitoring. Standardized tests for evaluating distinct neurocognitive domains include the Trail Making Test Parts A and B, Hopkins Verbal Learning Test-Revised, the Controlled Oral Word Association Test, the Digit Span, the Boston Naming Test, and the Wechsler Adult Intelligence Scale, 4th edition (17).
Acute radiation toxicity is often managed with corticosteroids. Corticosteroids may hasten improvement in some patients with early delayed encephalopathy. A slow corticosteroid taper and temporary return to higher doses in patients who were previously prescribed this for cerebral edema or other reasons may be necessary.
No treatment is necessary for either subacute myelopathy or transient brachial plexopathy, both of which are self-limited.
Delayed radiation injury is often irreversible and treatment is largely aimed at symptom control. The preventative strategies described in the Prevention section above are generally more successful than management strategies.
In many cases, radiation necrosis can be managed conservatively without intervention. In more severe cases, corticosteroids can be helpful in reducing the mass effect and associated deficits caused by edema and/or inflammation associated with cerebral radiation necrosis. Additional treatments to consider include bevacizumab, surgical resection, and laser-interstitial thermal therapy (LITT). Hyperbaric oxygen, high-dose multivitamins, anticoagulation, and antiplatelet therapy have not been established to be efficacious, but are utilized at times.
Bevacizumab has been utilized with success in select cases of radiation necrosis and is under further investigation as a treatment for radiation necrosis and radiation-induced optic neuropathy. Small prospective trials in addition to retrospective studies have found that anti-VEGF therapy offered symptomatic relief and radiographic response in patients with radiation necrosis and low risk of cerebral hemorrhage (43; 89). However, other studies have reported worsening symptoms with bevacizumab (79; 34); further investigations to establish the safety and optimal patient cohort are needed.
Resection of progressive space occupying radiation necrosis may be considered after failure of conservative therapy if there is evidence for persistent or progressive neurologic symptoms (35). Laser-interstitial thermal therapy (LITT) has been proposed as a minimally invasive alternative to open craniotomy (71; 32). This technique requires only a small burr hole to enable probe access to the brain and may be more tolerable for patients. However, systemic literature reviews suggest that bevacizumab may outperform LITT in managing radiation necrosis in terms of both clinical and radiographic outcomes (Gecici et 2024; 86).
A possible adjunct or alternative to corticosteroids for managing cerebral edema and radionecrosis is Boswellia serrata, an extract of the Indian frankincense plant. Small prospective studies, including one placebo-controlled, double-blind randomized clinical trial demonstrated reduction in cerebral edema with this dietary supplement (37). There may also be a role for Boswellia serrata in the management of radiation necrosis (85).
For patients with late cognitive impairment due to radiation, limited treatment options are available and of largely unproven benefit. Pharmacologic interventions include the Alzheimer drug donepezil (72) and stimulants like methylphenidate or modafinil. Cognitive rehabilitation and behavioral intervention are under investigation as treatment for late cognitive impairment (25; 17; 70; 36).
Most radiation-related tumors are histologically and clinically aggressive. Treatment does not differ from the standard of care for any CNS malignancy, though the risks of reirradiation and cumulative dose need to be considered.
Acute radiation toxicity is often self-limited, and corticosteroids may hasten improvement.
The outcome of treatment of radiation necrosis with steroids, bevacizumab, and other treatments is highly variable. Surgical resection is the most effective option and is used for salvage when other methods fail.
Delayed radiation injury is often irreversible, but treatment described in Management may help to address symptoms.
All contributors' financial relationships have been reviewed and mitigated to ensure that this and every other article is free from commercial bias.
Puyao Li MD
Dr. Li of University of Vermont Medical Center has no relevant financial relationships to disclose.
See ProfileAlexander Schrager
Mr. Schrager of Larner College of Medicine has no relevant financial relationships to disclose.
See ProfileRimas V Lukas MD
Dr. Lukas of Northwestern University Feinberg School of Medicine received honorariums from Novartis and Novocure for speaking engagements, honorariums from Cardinal Health, Novocure, and Merck for advisory board membership, and research support from BMS as principal investigator.
See ProfileNearly 3,000 illustrations, including video clips of neurologic disorders.
Every article is reviewed by our esteemed Editorial Board for accuracy and currency.
Full spectrum of neurology in 1,200 comprehensive articles.
Listen to MedLink on the go with Audio versions of each article.
MedLink®, LLC
3525 Del Mar Heights Rd, Ste 304
San Diego, CA 92130-2122
Toll Free (U.S. + Canada): 800-452-2400
US Number: +1-619-640-4660
Support: service@medlink.com
Editor: editor@medlink.com
ISSN: 2831-9125
Neuro-Oncology
Oct. 11, 2024
Neuro-Oncology
Oct. 03, 2024
Neuro-Oncology
Sep. 25, 2024
Neuro-Oncology
Sep. 19, 2024
Neuro-Oncology
Sep. 18, 2024
Neuro-Oncology
Sep. 18, 2024
Neuro-Oncology
Sep. 18, 2024
Neuro-Oncology
Sep. 18, 2024