Contrast-enhanced T2-fluid-attenuated inversion recovery magnetic resonance imaging for radiotherapy target delineation in brain metastases: A narrative review

Abstract

Brain metastases (BM) are the most common type of brain tumor, with diagnosis, pre-treatment evaluation, and follow-up relying on imaging. Radiotherapy, including whole brain radiotherapy and stereotactic radiosurgery, is a key treatment modality for BM, and its efficacy is closely related to the precision of target delineation, which depends on clear and accurate multimodal imaging. The target delineation for BM is typically performed using a fusion technique combining computed tomography and magnetic resonance imaging (MRI). In routine MRI, contrast-enhanced T1-weighted imaging (T1WI + C) is considered the basic sequence, but it has certain limitations, such as the interference with lesion signals adjacent to large vessels and the cerebral cortex caused by enhanced vascular signals. T2-fluid-attenuated inversion recovery (T2-FLAIR) sequence suppresses the signal of free water [cerebrospinal fluid (CSF)] while highlighting the signal of bound water (intracranial edema). Contrast-enhanced T2-FLAIR (T2-FLAIR + C) sequence further improves visualization, thereby addressing the limitations of T1WI + C. Specifically, T2-FLAIR + C effectively suppresses CSF signals while clearly delineating enhanced lesions, particularly those on the brain surface and adjacent to the cerebral cortex. Additionally, its heightened sensitivity to low concentrations of gadolinium allows for indirect assessment of the vascularity of brain and meningeal tumors. However, the use of MRI simulation in radiotherapy for BM is not widespread, and the application of the T2-FLAIR + C sequence is not yet common. This narrative review elaborates on the technical principles of the T2-FLAIR + C sequence and current guidelines for BM radiotherapy target delineation, summarizes its clinical applications in BM radiotherapy, and discusses existing evidence gaps and future research directions. Notably, direct evidence supporting the benefit of T2-FLAIR + C for contouring accuracy, dosimetric optimization, and patient clinical outcomes remains limited. This review aims to provide a comprehensive reference for further research and clinical application of T2-FLAIR + C in BM radiotherapy.

Key Words: Brain neoplasms; Neoplasm metastasis; Magnetic resonance imaging; Radiotherapy planning; Contrast-enhanced T2-fluid-attenuated inversion recovery; Magnetic resonance imaging simulation

Core Tip: For brainmetastasis delineation, enhanced T2-fluid-attenuated inversion recovery (T2-FLAIR) sequence can significantly enhance the accuracy of target contouring in condition of low-dose gadolinium-based contrast agents. Simultaneously, in consideration of the heterogeneity inside the tumor enhanced T2-FLAIR sequence can help to provide a more personalized treatment.

INTRODUCTION

Brain metastases (BM) are devastating intracranial complications of systemic malignancies, arising when tumor cells disseminate from their original site to the brain through hematogenous or lymphatic pathways[1]. The pathological mechanism of BM is complex. Tumor cells must first break through the basement membrane of the primary site and enter the bloodstream, then stagnate and adhere to the microvasculature in the brain. Subsequently, they extravasate from the vascular wall and proliferate within brain tissue to form metastatic foci[2,3]. This process results in the development of intracranial tumors. Magnetic resonance imaging (MRI) remains the gold standard for BM detection. On MRI, BM can be categorized into brain parenchymal metastasis and meningeal metastasis, with common clinical symptoms including headache, vomiting, blurred vision, hemiplegia, and dysphasia, which are clinical manifestations and do not define metastasis categories based on imaging appearance[4,5]. The brain parenchyma is the most frequent intracranial metastatic site[6,7]. Notably, different MRI sequences have varying effects on the detection and characterization of BM. On conventional T2-weighted imaging, BM typically appears as hyperintense lesions, often surrounded by edema[8]. However, when lesions are small or located near cerebrospinal fluid (CSF), the high signal intensity of CSF may obscure the lesion, making it difficult to visualize clearly. In contrast, T2-fluid-attenuated inversion recovery (T2-FLAIR) sequence effectively suppresses CSF signals and enhances the visualization of intracranial lesions and associated edema, which has been widely applied in central nervous system diseases[9]. Additionally, conventional contrast-enhanced T1-weighted imaging (T1WI + C) in MRI may be compromised by vascular enhancement when evaluating lesions adjacent to CSF or cerebral cortex[10]. In comparison, contrast-enhanced T2-FLAIR (T2-FLAIR + C) sequence demonstrates advantages by combining T2 and partial T1 weighting characteristics. This sequence displays intracranial lesions as hyperintense signals while showing perilesional edema as slightly hyperintense, thereby improving differentiation between pathological and edematous regions[11]. Moreover, it utilizes flow-void effect to minimize vascular signal interference, allowing clearer visualization of lesions near CSF or cerebral cortex.

For this narrative review, a targeted literature search was conducted in PubMed/MEDLINE, EMBASE, Web of Science, and the Cochrane Library, covering literature published from January 2010 to June 2026. Search terms included "T2-FLAIR", "contrast-enhanced T2-FLAIR", " T2-FLAIR + C ", "brain metastases", "radiotherapy", "target delineation", "MRI simulation", "stereotactic radiosurgery", "whole brain radiotherapy" and their corresponding Mesh terms. Eligible publications included English-language peer-reviewed clinical trials and observational studies focused on BM and related MRI sequences. This search was selectively focused on topics relevant to BM radiotherapy target delineation, and does not meet the methodological standards of a systematic review. Limitations include language restriction and potential omission of non-English and grey literature. Radiotherapy is an important part of BM treatment[12], with strategies including neoadjuvant radiotherapy, adjuvant radiotherapy, and definitive radiotherapy depending on the specific clinical BM conditions. Stereotactic radiosurgery (SRS) is one of the main radiotherapy modalities for BM[13], particularly for patients with a limited number of metastases, characterized by high-energy, high-dose, highly focused, and precise radiation[14,15]. Accurate target delineation fundamentally determines radiotherapeutic efficacy. Current guidelines recommend computed tomography (CT)/MRI simulation (MRI-sim) image registration for optimal target definition[16]. MRI-sim sequences differ from conventional MRI diagnostic sequences in terms of scanning parameters; the former focuses on accurate tumor boundary delineation to ensure treatment precision, while the latter emphasizes tumor detection for diagnosis[17]. In addition, the brain sequence in MRI-sim uses 3D imaging technology. Compared to conventional 2D fluid-attenuated inversion recovery, 3D fluid-attenuated inversion recovery (3D-FLAIR) sequence achieves thinner slice thickness (≤ 1 mm) scans through variable flip-angle fast spin-echo techniques, effectively reducing CSF pulsation artifacts. Diagnostic imaging studies have reported that 3D-FLAIR sequence improves the detection rate of small lesions (diameter < 5 mm) by approximately 20% under low-dose contrast conditions compared with conventional T1WI + C; however, this finding has not been validated in radiotherapy target delineation scenarios[18]. The Clinical Practice Guidelines for the Management of Brain Metastases from Non-small Cell Lung Cancer with Actionable Gene Alterations in China (2025 Edition) highly recommend "T2-FLAIR + C sequence with slice thickness ≤ 1.5 mm" as the preferred imaging for BM diagnosis, emphasizing its clinical importance[19]. However, the application of T2-FLAIR + C sequence is not yet common, and the use of MRI-sim in radiotherapy for BM is also not widespread. It is essential to integrate T2-FLAIR + C sequence into the conventional enhanced MRI-sim sequence before the BM radiotherapy treatment to improve target delineation accuracy.

This narrative review synthesized current evidence on the technical principles of T2-FLAIR + C, its applications in radiotherapy planning, and future directions for optimizing BM management. By integrating available evidence with clinical practice insights, we aimed to summarize the current state of this imaging modality in radiation oncology and provide a reference for further research and clinical application in BM radiotherapy.

TECHNICAL CHARACTERISTICS OF T2-FLAIR + C SEQUENCE

Mechanism and advantages of T2-FLAIR + C sequence

The enhancement mechanism of contrast agents in MRI differs from that of iodinated contrast agents in CT. Contrast agents in MRI indirectly enhance imaging by altering the surrounding local magnetic field, whereas in CT, iodinated contrast agents enhance images by directly increasing X-ray attenuation. In MRI, the enhancement mechanism involves shortening T1 relaxation time of tissues. This is achieved through paramagnetic effects, such as gadolinium, manganese, and deoxyhemoglobin, and hydration layer effects, such as proteins and calcium ions. The most commonly used paramagnetic substances in MRI are gadolinium-based contrast agents (GBCA)[20]. They accelerate longitudinal relaxation and increase the magnetization vector recovery, resulting in higher signal intensity on T1-weighted images. This is particularly evident in post-enhancement images, where vessels show increased signal intensity[21]. In contrast, CT imaging relies on inherent tissue contrast, which can be limited due to the similar densities of tissues like muscle, blood vessels, fat, and solid organs. These tissues exhibit similar X-ray attenuation, leading to a relatively uniform gray scale in CT images. Iodine-based contrast agents enhance CT images by increasing the density of tissues, thereby creating differences in X-ray penetration and improving the visibility of tissues and organs[22,23]. Most malignant tumors have a compromised blood-brain-barrier (BBB) phenomenon, leading to increased vascular permeability and subsequent leakage of contrast agents. As a result, tumors show significant enhancement on CT[24].

The T2 FLAIR sequence is a T2-weighted inversion recovery sequence. Following a 180° inversion pulse, the magnetization vectors of all tissues are inverted to the -Z axis direction. Upon cessation of the inversion pulse, these vectors begin to relax. However, the inversion time (TI) is insufficient for all tissues to complete longitudinal relaxation, resulting in some degree of longitudinal relaxation contrast[25]. Therefore, the shortening of T1 relaxation time induced by paramagnetic contrast agents also increases signal intensity on T2-FLAIR images. Studies have shown that about 82.76% of BM patients exhibit significant enhancement on T2-FLAIR + C sequence[26]. This sequence features long repetition time (TR), echo time, and TI, with TI suppressing CSF signals containing free water to low signal intensity. The extended TR allows cortical vessels and venous sinuses to display flow-void effect[27], helping to differentiate abnormal meningeal enhancement or lesions near the meninx from the normal enhancement of cortical blood vessels seen in conventional T1-weighted imaging (T1WI) sequences[28]. T2-FLAIR + C sequence is not only useful for tumor localization but also for assessing the biological behavior of tumors, such as determining tumor vascular permeability through enhancement level, which can provide additional information for treatment planning in radiotherapy[29].

Characteristics of T2-FLAIR signal enhancement

The enhancement level of lesions on T2-FLAIR + C sequence varies with different doses of the contrast agent[30,31]. Studies have shown that T2-FLAIR + C sequence is approximately 10 times more sensitive to low-dose contrast agents than conventional T1WI + C. This heightened sensitivity allows T2-FLAIR + C sequence to more effectively capture the shortening of T1 relaxation time, particularly when the dose of the contrast agent is low. Consequently, lesions that exhibit faint enhancement on conventional T1WI + C sequence can be more clearly displayed on T2-FLAIR + C sequence, thereby improving the detection rate of small lesions with low enhancement levels[31,32]. However, there is no direct clinical evidence confirming that radiotherapy plans incorporating T2-FLAIR + C can improve local control of BM; this hypothesis remains to be validated in large-scale prospective clinical studies.

In vitro phantom study further demonstrated that the concentration of GBCA required to generate reliable signal intensities on T2-FLAIR + C sequence was approximately 0.12-0.48 mmol/L, which is one-eighth to one-half lower than the concentration needed to achieve comparable signal intensity on T1WI + C[33]. Specifically, half-dose and standard-dose contrast agents were administered, and the contrast-to-noise ratio and percentage increase of both sequences were calculated under different dosage conditions. Variance analysis was subsequently used to compare the objective measurement results between groups. The findings revealed that on T2-FLAIR + C sequences with half-dose contrast agent, the enhancement level of BM was comparable to that achieved with standard-dose on T1WI + C sequences[33]. This not only compensates for the limitations of conventional T1WI + C sequences in detecting small metastatic tumors but also benefits patients at potential risk of allergic reactions to GBCA.

Moreover, studies have indicated that after multiple enhanced scans using GBCA, deposition of the contrast agent can occur in the basal ganglia of patients[34,35]. This deposition poses a potential risk to the patients, thereby highlighting the necessity and trend of reducing GBCA doses in future applications. This characteristic facilitates the differentiation between normal vessels and abnormal lesions, eliminates interference from surface vascular shadows, and allows for a clear display of tumor size and boundaries[36]. Some studies have suggested that radiotherapy plans incorporating T2-FLAIR + C may be associated with improved local control, though these findings require validation in larger prospective studies[37].

Enhancement of intracranial normal tissues in T2-FLAIR + C

T2-FLAIR + C sequence enhances normal intracranial tissues such as the choroid plexus, pituitary gland, cavernous sinus and its contents, dura mater, venous sinuses and pineal gland, etc.[38]. For example, the choroid plexus lacks a complete BBB effect, and the fenestrated endothelial cells of the capillaries allow contrast agents to seep through, resulting in enhanced signal on T2-FLAIR + C sequence[39,40]. The enhancement level of different parts of the dura mater also varies on T2-FLAIR + C sequence[41]. These physiological enhancements represent potential diagnostic pitfalls and may lead to false-positive interpretation if not recognized.

DEVELOPMENT OF RADIOTHERAPY AND CT/MRI-SIM SYSTEMS

Radiotherapy refers to the use of high-energy radiation to damage the DNA of cancer cells, rendering them incapable of division and replication, thus reducing or eliminating tumor tissue to achieve cancer control[42-45].

The accuracy of the information collected by the CT simulation (CT-sim) and localization system directly impacts the precision of subsequent radiotherapy delivery to tumors. Compared to CT imaging, MRI provides higher resolution images, enhancing tumor identification, especially in the brain, central nervous system, and abdominal-pelvic soft tissues[46,47]. Unlike diagnostic MRI, which uses a curved couch, small apertures, and specific MRI coils, MRI-sim uses a flat couch, larger apertures, and specialized coils for radiotherapy. Diagnostic MRI cannot accommodate the fixed immobilization devices used in CT-sim, such as the head-neck-shoulder mold for cranial tumors, leading to potential discrepancies when registering MRI and CT images[48-51]. MRI-sim overcomes these discrepancies by ensuring consistent patient positioning for both CT-sim and MRI-sim, thereby achieving higher immobilization accuracy. Furthermore, MRI-sim performs daily geometric distortion quality assurance (QA) to verify that the geometric distortions was within the acceptable threshold, thus guaranteeing the accuracy of the tumor's geometry. The German Working Group on Radiotherapy Imaging has published guidelines emphasising that accurate MRI-sim is the foundation of high-precision SRS and fractionated stereotactic radiotherapy, necessitating radiotherapy-optimised MRI protocols and clear quality criteria to ensure geometrically accurate images for treatment planning[52]. Similarly, the AAPM Task Group 284 report provides a standardised framework for assessing geometric distortion in MRI. In routine practice, QA programmes involve periodic geometric accuracy tests using dedicated phantoms and evaluation of distortion across different sequences. It has been shown that the average distortion for all 3D sequences is around 0.4-0.5 mm, although time-of-flight angiography sequences may exhibit maximum distortions up to 3.3 mm. Geometric distortion is a critical factor affecting target localization accuracy in MRI-based radiotherapy planning. It arises primarily from gradient nonlinearity, main magnetic field inhomogeneity, susceptibility effects, and chemical shift, and tends to increase with distance from the isocenter. In the context of BM, Ohira et al[53] demonstrated that for lesions located within 30 mm of the MRI isocentre, the median geometric distortion was approximately 0.1 mm; however, when the distance exceeded 60 mm, the median distortion increased significantly (0.23-0.57 mm even after distortion correction). More importantly, the same study found that without distortion correction, the dose covering 98% of the target volume could fall below prescription by more than 5% for targets situated 48 mm from the isocenter, with a maximum deficit of 26.5%. In the context of BM, Ohira et al[53] demonstrated that for lesions located within 30 mm of the MRI isocenter, the median geometric distortion was approximately 0.1 mm; however, when the distance exceeded 60 mm, the median distortion increased significantly (0.23-0.57 mm even after distortion correction). Scanning parameters in MRI-sim, including slice thickness, scanning orientation, and spacing, are aligned with those of CT-sim to facilitate a more precise tumor localization during CT/MRI image registration. Current researches focus on using MRI-sim directly for target delineation and converting MRI images into CT electron density maps for dose calculation, thereby simplifying radiotherapy treatment planning and workflow[52-56]. MRI-only planning has gained increasing attention with advances in deep learning, enabling the generation of synthetic CT from MRI to eliminate the need for CT-sim. The advantages of an MRI-only workflow include removal of CT-MRI registration errors, reduced patient radiation exposure, and streamlined treatment pathway. The importance of MRI in radiotherapy cannot be overstated, and the progress of T2-FLAIR + C sequence for delineating target areas in BM radiotherapy has significant potential for routine clinical practice.

BASIC PRINCIPLES AND IMPLEMENTATION OF TARGET DELINEATION FOR BM RADIOTHERAPY

Target delineation in radiotherapy is a critical step in the treatment of BM. The basic principle is to accurately identify the tumor areas that require irradiation while minimizing radiation damage to the surrounding normal tissues[57,58]. To achieve this, the boundaries of the tumor should be precisely identified based on multimodal images such as MRI, CT, or positron emission tomography-CT (PET-CT) scans. Additionally, the biological behavior of the tumor must be considered. For highly invasive BM, the target area may need to be appropriately expanded to cover potential areas of tumor infiltration[59,60]. In practical implementation, there are standardized definitions for target volumes and dose descriptions. The Gross Tumor Volume (GTV) is identified through clinical validation and diagnostic imaging, representing the visible range of malignant lesions with certain shape and size. The Clinical Target Volume takes into account the subclinical infiltration range based on the GTV, while the Planning Target Volume (PTV) further considers factors such as setup errors to ensure that the tumor receives a sufficient radiation dose throughout the treatment process[61,62]. Studies have shown that on T2-FLAIR + C sequence of MRI, the approximate tumor boundaries can be initially determined based on the extent of lesion enhancement[63]. Radiotherapy for BM is implemented following ASTRO guidelines, which recommend either SRS or whole brain radiotherapy (WBRT) technique. Compared to traditional techniques, SRS is characterized by high single-dose, fewer treatment fractions, and image-guided precise radiotherapy, which significantly reduces damage to normal tissue and minimizes side effects of radiotherapy[64-66]. SRS is recommended for patients with an Eastern Cooperative Oncology Group performance status of 0-2 and up to 4 intact BM[67-72]. For patients with favorable prognosis, estimated using a validated brain metastasis prognostic index, and BM ineligible for surgery or SRS, WBRT (e.g., 3000 cGy in 10 fractions) is recommended as the primary treatment[73-76]. Some guidelines in BM radiotherapy, whether using SRS or WBRT, are addressed according to the ASTRO (Table 1)[77].

Table 1 An ASTRO clinical practice guideline for brain metastases radiotherapy.

KQ	Population	Intervention	Comparator	Outcomes
1	What are the indications for SRS alone for patients with intact BM?
	Patients with intact BM	Observation WBRT	SRS	Intracranial control; progression-free survival; overall survival; neurocognitive function; patient-reported outcomes
2	What are the indications for observation, preoperative SRS or WBRT in patients with resected BM?
	Patients with resected BM	Observation WBRT	SRS	Intracranial control; progression-free survival; overall survival; neurocognitive function; patient-reported outcomes
3	What are the indications for WBRT for patients with intact BM?
	Patients with intact BM	Observation WBRT	Conventional WBRT; HA-WBRT; HA-WBRT plus memantine	Intracranial control; progression-free survival; overall survival; neurocognitive function; patient-reported outcomes
4	What are the risks of symptomatic radionecrosis with WBRT and/or SRS for patients with BM?
	Patients with BM	WBRT	SRS	Symptomatic radionecrosis; other adverse effects

BM: Brain metastases; KQ: Key question; SRS: Stereotactic radiosurgery; WBRT: Whole-brain radiation therapy; HA-WBRT: Hippocampal-avoidance whole-brain radiation therapy.

APPLICATION OF T2-FLAIR + C SEQUENCE IN RADIOTHERAPY FOR BM

Meningeal metastasis

MRI is a crucial tool in disease diagnosis and scientific research, and advancements in MRI technology have led to more frequent detection of meningeal metastases[78,79]. A hallmark imaging feature of meningeal metastasis is the dural tail sign[80,81]. The routes of metastasis include hematogenous spread, lymphatic spread, CSF seeding, and direct invasion[82], with hematogenous dissemination being the most common. Malignant tumors from various parts of the body can metastasize to the meninges[83]. Meningeal metastasis can be classified into leptomeningeal metastasis and dural metastasis based on the site of metastasis.

Leptomeningeal metastasis, which results from the invasion of cancer cells into the pia mater and subarachnoid space[84], manifests as small nodular or linear lesions tightly adhering to the brain surface and can extend into sulci and cisterns. This type of metastasis is primarily located in the pia mater, subarachnoid space, ependyma, and ventricular walls, often accompanied by ventricular enlargement. The most common primary cancers causing leptomeningeal metastasis are breast cancer, lung cancer, and melanoma[85,86]. On T2-FLAIR sequence, leptomeningeal metastasis, particularly when involving the pia mater and subarachnoid space, show multiple linear high signal lesions or nodular shadows along the sulci[86]. Ventricular metastasis, which is relatively rare, is often associated with leptomeningeal and subarachnoid metastases. It appears as band-like or nodular abnormal signals along the ventricular walls with significant enhancement, and the ventricular system is often expanded to varying enhancement levels. This may result in secondary periventricular interstitial brain edema, which appears as diffuse high signals in the periventricular white matter on FLAIR sequences[87], with no enhancement on enhanced scans.

Dural metastasis can occur due to skull metastasis involving the dura or through hematogenous spread[88]. It typically presents as nodular or thickened segmental linear lesions on the brain surface, tightly adherent to the inner skull layer and following the course of the falx cerebri and tentorium cerebelli, without extending into sulci or cisterns. These lesions can be seen in consecutive layers[89]. The adjacent skull may show metastasis or appear normal[90]. In areas near the skull, such as the posterior fossa, T1WI + C is easily affected by magnetic susceptibility artifacts, while T2-FLAIR + C sequence is more stable. Furthermore, T2-FLAIR + C sequences display the extent, number, and enhancement level of meningeal metastases more clearly than conventional T1WI + C sequences[32,91].

Brain parenchymal metastasis

Brain parenchymal metastasis most commonly occurs in the cerebral hemispheres, followed by the cerebellum and brainstem[92]. These metastases are more frequent in the cerebral hemispheres than in the cerebellum, with solitary lesions more often found in the supratentorial region than the subtentorium[93]. The primary sources of brain parenchymal metastases are mainly lung cancer and breast cancer[4]. Hematogenous spread is the predominant pathway, with tumor emboli often blocking vessels at the gray-white matter junction due to sudden narrowing of blood vessels, leading to metastatic lesions. Therefore, the gray-white matter junction is the most common site for metastasis and is prone to multiple metastatic lesions.

On imaging, typical brain parenchymal metastases often appear as round or oval lesions with clear boundaries, varying in size[94,95]. On unenhanced T1WI sequences, these lesions usually present as slightly low signal or isointense, with high signal intensity possible in cases of hemorrhage or melanoma metastasis. On T2-FLAIR sequence, they typically appear slightly hyperintense, though isointense, often with central non-enhancing necrosis. T1WI + C sequence shows significant lesion enhancement, often with central areas exhibiting non-enhancing necrosis. Surrounding edema, a common cause of clinical symptoms[96], is well visualized on T2-FLAIR sequences[97,98]. T2-FLAIR + C sequence further enhances differentiation between edema and the lesion areas by demonstrating varying signal intensities[99]. Significant edema can cause compression of surrounding tissues, midline shift, and any signs of brain herniation, which require urgent clinical intervention[100].

Inclusion of T2-FLAIR + C sequence effectively reduces misdiagnosis rates and provides valuable guidance for radiotherapy planning, ensuring the accuracy of target delineation. Advantages and limitations of different imaging modalities in BM target delineation are shown in Table 2.

Table 2 Comparison of different imaging modalities in brain metastases target delineation.

Imaging modality	Advantages	Limitations	Clinical applications in BM radiotherapy
T1WI + C	Standard reference; excellent for enhancing lesions; widely available	Vascular enhancement can obscure cortical/CSF-adjacent lesions; signal interference from blood vessels	Primary sequence for GTV delineation; standard for response assessment
T2-FLAIR	Suppresses CSF signal; excellent for edema visualization; detects non-enhancing tumor	No contrast enhancement; may miss small lesions	Assessing peritumoral edema; visualizing non-enhancing tumor components
T2-FLAIR + C	CSF suppression; superior detection of cortical/meningeal lesions; high sensitivity to low-dose contrast; reduces vascular interference	Variable availability; technical standardization lacking; limited evidence base for target delineation	Delineating lesions near CSF/cortex; detecting leptomeningeal disease; possible role in target delineation
CT	Fast; widely available; excellent for bone anatomy; essential for dose calculation	Poor soft tissue contrast; low sensitivity for small metastases	Simulation and localization; electron density mapping for dose calculation
PET-CT	Metabolic information; distinguishes active tumor from necrosis/edema	Limited spatial resolution; higher cost; radiation exposure	Problem-solving in equivocal cases; detecting active disease
DCE-MRI	Assesses vascular permeability; biological information	Longer acquisition; specialized processing; not standard in routine practice	Research applications; understanding tumor biology
DSC-MRI	Assesses perfusion and angiogenesis; helps differentiate RN from TP	Requires specialized software; prone to artifacts	Differentiating radiation necrosis from tumor progression

T1WI + C: Contrast-enhanced T1-weighted imaging; T2-FLAIR: T2-fluid-attenuated inversion recovery; T2-FLAIR + C: Contrast-enhanced T2-fluid-attenuated inversion recovery; CT: Computed tomography; PET-CT: Positron emission tomography-computed tomography; DCE-MRI: Dynamic contrast-enhanced magnetic resonance imaging; DSC-MRI: Dynamic susceptibility contrast magnetic resonance imaging; BM: Brain metastases; CSF: Cerebrospinal fluid; GTV: Gross Tumor Volume; RN: Radiation necrosis; TP: Tumor progression.

Summary and outlook

T2-FLAIR + C sequence has unique advantages for BM diagnosis compared to other MRI sequences. Relative to CT, T2-FLAIR + C sequence is more sensitive in detecting brain parenchymal lesions, particularly small metastatic lesions that are difficult to identify with CT. Unlike the T1WI + C, T2-FLAIR + C sequence effectively suppresses CSF signals while enhancing lesion visibility, making it more sensitive for detecting lesions on the meninges and cerebral cortex. Although PET-CT excels in detecting tumor metabolic activity[101,102], T2-FLAIR + C provides clearer anatomical structures of tumors and better delineation of the signal differences between tumors and surrounding brain tissue. Recently, artificial intelligence driven target delineation in BM develops rapidly. Precision tumor segmentation and radiomics and outcome prediction were all published in variety of references[103-105]. Additionally, the use of low-dose contrast agents in T2-FLAIR + C sequence enhances lesion visualization while minimizing the risk of gadolinium deposition in BM patients.

Radiotherapy can be used as a standalone curative treatment or in combination with surgery, chemotherapy, and palliative care[106-111]. The evolution of radiotherapy for BM has progressed through several stages. Initially, conventional techniques involved large irradiation fields to cover potential metastatic areas, causing significant damage to surrounding normal brain tissue[112,113]. With advancements in technology, SRS emerged, allowing precise targeting and highly focused and high-dose irradiation of tumors while minimizing the radiation exposure to surrounding normal brain tissues, thereby reducing toxicity[72,114]. The widespread use of T2-FLAIR + C sequence in BM diagnosis has led to its gradual incorporation into radiotherapy workflows. Additionally, T2-FLAIR + C sequence is utilized in post-treatment follow-up to assess therapeutic efficacy, monitor tumor recurrence, and track disease progression. T2-FLAIR + C sequence thus holds significant value in BM diagnosis and management. They enhance the detection rates of BM[115,116] and provide valuable insights into tumor biology. Specifically, T2-FLAIR + C sequence facilitates tumor recognition and evaluates the biological behavior of tumors by analyzing the enhancement level of metastatic lesions. Studies have shown that the enhancement level of BM on T2-FLAIR + C sequence is correlated with the vascular permeability parameters in dynamic contrast-enhanced MRI, with higher enhancement levels indicating lower Ktrans and Kep values. This correlation provides a deeper understanding of the biological characteristics of tumors[117,118], thereby supporting the development of more effective radiotherapy strategies.

Critical appraisal and controversies

Strengths of T2-FLAIR + C: (1) Excellent detection performance for leptomeningeal metastases and superficial cortical lesions; (2) Ultra-high sensitivity to low-dose GBCA; (3) Flow void effect suppresses vascular interference, clearly separating lesions from normal cerebral vessels; and (4) Simultaneous display of metastatic lesions and peritumoral vasogenic edema, facilitating integrated target delineation.

Comprehensive discussion of limitations and diagnostic pitfalls: (1) Inconsistent lesion enhancement manifestations under different magnetic field strengths and MRI equipment[119]; (2) Lack of unified standardized scanning protocols for radiotherapy simulation scenarios; and (3) False positive enhancement of normal intracranial tissues (choroid plexus, pituitary gland, dura mater, venous sinus and pineal gland) leading to misdiagnosis risk.

Core unresolved academic controversies: (1) Whether T2-FLAIR + C serves as a supplementary sequence or a partial substitute for routine T1WI + C; (2) Optimal gadolinium dosage strategy (standard dose vs half low dose); (3) Clinical significance of lesions only visible on T2-FLAIR + C but negative on T1WI + C; and (4) Whether T2-FLAIR + C enhancement signals truly reflect tumor vascular permeability and biological activity.

Compare the differentiation ability of T2-FLAIR + C and dynamic susceptibility contrast-MRI for radiation necrosis (RN), and discuss the limitation that T2-FLAIR + C lacks unified quantitative parametric evaluation standards at present.

FUTURE DEVELOPMENT DIRECTIONS OF T2-FLAIR + C IN BM TREATMENT

Nowadays, the potential of T2-FLAIR + C sequence continues to be explored. In terms of equipment, the magnetic field strength of MRI devices is constantly increasing, moving from the commonly used 1.5 T and 3.0 T to higher field strengths. This further improves image resolution and signal-to-noise ratio, enhancing the effectiveness of T2-FLAIR + C sequence in displaying the details of BM[99,120]. In terms of image processing, new image reconstruction algorithms and post-processing techniques are emerging. For example, advanced denoising algorithms reduce image noise without losing detail, thus improving image quality. Image registration technology also allows T2-FLAIR + C sequence to be better integrated with other imaging modalities (such as PET-CT), providing more comprehensive information for evaluating BM[121,122]. In terms of assessment for radiotherapy efficacy, clinical observations show that some patients exhibit increased enhancement areas or new enhancing lesions on follow-up MRI after radiotherapy or chemotherapy. Distinguishing RN from tumor progression (TP) on conventional MRI is challenging[123,124]. Recent studies have shown that T2-FLAIR + C sequence effectively differentiates RN from TP after radiotherapy for BM, with discrimination capabilities comparable to those of Dynamic Susceptibility Contrast perfusion imaging[125,126]. These findings indicate the diagnostic potential of T2-FLAIR + C sequence for diagnosing RN and present opportunities for developing new qualitative and quantitative methodologies to exploit this contrast mechanism. In Heyn et al’s study[123], T2-FLAIR + C sequence surprisingly offers the ability to qualitatively inspect images or produce parametric maps at a spatial resolution equivalent to that of conventional MRI. While a rigorous analysis of these images or maps was not undertaken, spatial heterogeneity in T2-FLAIR + C sequence was observed with similarities to the heterogeneity observed with regional cerebral blood volume maps[125]. For clinicians, T2-FLAIR + C sequence, being more convenient, repeatable, and relatively widely accessible, is expected to become an important imaging tool for assessing post-radiotherapy lesions. This can help to reduce unnecessary invasive procedures and guide more precise subsequent treatment plans. Summary of key clinical studies of T2-FLAIR + C sequence in BM is shown in Table 3.

Table 3 Summary of key clinical studies of contrast-enhanced T2-fluid-attenuated inversion recovery sequence in brain metastases.

Ref.	Population	Sample size	Contrast agent/dose	Acquisition setting	Comparator	Primary outcome	Key limitations
Jin et al[33]	Patients with suspected BM	82	Half-dose	Diagnostic	Standard-dose T1WI + C	Lesion detection rate; contrast-to-noise ratio comparison	Single-center; no radiotherapy contouring endpoint
Ahn et al[31]	Patients with small BM	121	Double-dose	Diagnostic	Conventional 3D T1WI + C	Detection sensitivity for small BM	Diagnostic cohort only; no radiotherapy application
Sun et al[18]	Lung cancer patients with BM	156	Standard-dose	Diagnostic	T1WI + C alone	Overall lesion detection rate	Diagnostic study; no target delineation analysis
Heyn et al[123]	BM patients post-radiotherapy	98	Standard-dose	Diagnostic follow-up	DSC-MRI	Differentiation of radiation necrosis vs tumor progression	Retrospective; single-center
Ohira et al[53]	BM patients undergoing SRS	41	Standard-dose	Radiotherapy simulation	CT simulation	Geometric distortion and dosimetric impact	Focused on distortion, not contour accuracy

All available evidence is derived from diagnostic imaging or geometric distortion studies. No prospective study has directly evaluated the impact of contrast-enhanced T2-fluid-attenuated inversion recovery on Gross Tumor Volume contouring accuracy, interobserver variability, dosimetric outcomes, or patient survival. T1WI + C: Contrast-enhanced T1-weighted imaging; CT: Computed tomography; DSC-MRI: Dynamic susceptibility contrast magnetic resonance imaging; BM: Brain metastases; SRS: Stereotactic radiosurgery.

Moreover, T2-FLAIR + C sequence may be combined with emerging treatment technologies such as immunotherapy and gene therapy[29,127]. By evaluating the tumor microenvironment through T2-FLAIR + C sequence, it can provide more precise target information for immunotherapy and gene therapy, optimizing combined treatments and improving the treatment effectiveness of BM.

MRI-guided radiotherapy represents a significant advancement in the field of radiotherapy in recent years[128,129]. The integration of MRI devices with linear accelerators (linacs) enables real-time visualization of tumor position and morphological changes during treatment, facilitating online adaptive radiotherapy. This approach reduces radiation exposure to surrounding critical structures, achieving dynamically adjusted and more precise radiotherapy[128,130,131]. For tumors with significant motion, MRI-linacs can implement breath-hold gating or real-time tracking[131,132]. Compared to cone beam CT, MRI-guided radiotherapy offers distinct advantages, including high resolution, superior soft tissue contrast, and the absence of ionizing radiation[133,134].

In BM, where rapid TP and surrounding edema are common, MRI-guided radiotherapy is particularly advantageous. During treatment, changes in tumor geometry and position frequently occur due to factors such as edema resolution and treatment response, leading to significant shifts compared to pre-treatment imaging[135]. Research has shown that in long fractionated radiotherapy for BM, the dose coverage of the PTV decreases by 34.8% compared to pre-treatment levels. Performing MRI during treatment helps optimize dose distribution and improve the precision of radiotherapy[136]. Furthermore, MRI-linacs can be used for WBRT in BM treatment, with treatment plan quality and delivery accuracy comparable to conventional methods[137].

Given that traditional CT-based planning yields reasonable dosimetric results, MRI-guided radiotherapy for hippocampal-sparing WBRT offers possibilities for radiomics analysis and adaptive treatment strategies based on disease response (i.e., enhancement)[137]. MRI-guided radiotherapy allows for the acquisition of T1, T2-weighted, and other multi-functional sequences before each treatment session. For BM, using low-dose GBCA in T2-FLAIR + C sequence could better visualize edema and tumor geometry changes during treatment, enabling real-time adjustments to the target volume. This significantly improves the alignment between the radiation field and the tumor, resulting in better intracranial lesion control[138,139].

CONCLUSION

T2-FLAIR + C sequence holds great potential for personalized radiotherapy treatment by assessing tumor biological characteristics, such as vascular permeability and metabolism[139-141], providing the basis for creating personalized radiotherapy treatment plans for each patient. For instance, for tumors with high vascular permeability, the prescription dose and fractionation scheme can be appropriately adjusted to improve treatment efficacy. Additionally, by integrating the patient's genetic testing results, T2-FLAIR + C sequence can offer more precise guidance for targeted therapy and immunotherapy. Specifically, if a patient's tumor has specific genetic mutations, T2-FLAIR + C sequence can identify the distribution and active regions of the tumor, guide the precise delivery of targeted drugs and enable comprehensive, personalized treatment[142].