BT-RADS score
The effective use of BT-RADS by radiologists requires the fulfillment of certain technical and clinical prerequisites:
- Brain MRI with fluid-attenuated inversion recovery (FLAIR) and contrast-enhanced T1-weighted images
- Clinical indications of post-treatment surveillance in patients with intra-axial primary brain cancers
- Knowledge of the surgical, radiation, and medication history
BT-RADS employs a standardized numerical scoring system (0-4), with certain categories further subdivided (1a/1b, 3a/3b/3c) for precise assessment. Each score is directly linked to specific management recommendations
- BT-RADS 0: baseline study (initial diagnostic MRI or most recent MRI after surgery), non-tumor findings obscure diagnosis (e.g. infection), or non-diagnostic study or otherwise unable to classify
Fig 2: BT-RADS 0 example. A and C are fluid-attenuated inversion recovery (FLAIR) images, while B and D are contrast-enhanced T1-weighted images from two consecutive magnetic resonance imaging (MRI) scans. Patient history includes left frontal lobe glioblastoma resection in August 2020, followed by radiation therapy and a subsequent surgical resection of tumor recurrence in November 2021. The patient additionally commenced treatment with temozolomide. Since the December 2021 MRI is the first after the new surgery, this will serve as a baseline for future follow-ups (BT-RADS 0). - BT-RADS 1a: reflects decreasing tumor burden and/or improving treatment effect
Fig 3: BT-RADS 1a example. A and C are fluid-attenuated inversion recovery (FLAIR) images, while B and D are contrast-enhanced T1-weighted images from two consecutive magnetic resonance imaging (MRI) scans. Patient history includes partial resection of a frontal oligodendroglioma in November 2020, followed by radiation therapy completed in February 2021. The April 2021 MRI reveals a reduction in FLAIR hyperintensity, mass effect, and contrast-enhancement along the surgical cavity margins, indicating a favorable response to treatment without bevacizumab or high-dose steroids (BT-RADS 1a).
Fig 4: BT-RADS 1a example. A and C are fluid-attenuated inversion recovery (FLAIR) images, while B and D are contrast-enhanced T1-weighted images from two magnetic resonance imaging (MRI) scans of the same patient in Figure 5 and 12. Patient history includes resection of left parietal lobe recurrent glioblastoma in December 2015 and in September 2016, followed by radiation therapy completed in November 2016. The patient additionally commenced treatment with bevacizumab in November 2016. The February 2018 MRI reveals a reduction in contrast-enhancement along the surgical cavity margins. Since this is not the first examination after commencing bevacizumab treatment (see Figure 5), category 1a can be assigned. - BT-RADS 1b: reflects improvement from increasing doses of steroids or initiating bevacizumab
Fig 5: BT-RADS 1b example. A and C are fluid-attenuated inversion recovery (FLAIR) images, while B and D are contrast-enhanced T1-weighted images from two consecutive magnetic resonance imaging (MRI) scans of the same patient in Figure 4 and 12. Patient history includes resection of left parietal lobe recurrent glioblastoma in December 2015 and in September 2016, followed by radiation therapy completed in November 2016. The patient additionally commenced treatment with bevacizumab in November 2016. The February 2017 MRI reveals a reduction in FLAIR hyperintensity, mass effect, and especially contrast-enhancement along the surgical cavity margins, potentially related to the antiangiogenic effect of bevacizumab (BT-RADS 1b). - BT-RADS 2: reflects stable disease
Fig 6: BT-RADS 2 example. A and C are fluid-attenuated inversion recovery (FLAIR) images, while B and D are contrast-enhanced T1-weighted images from two consecutive magnetic resonance imaging (MRI) scans. Patient history includes resection of right temporal lobe astrocytoma in April 2023, followed by radiation therapy completed in July 2023. The patient additionally commenced treatment with temozolomide. The March 2024 MRI demonstrates a substantially stable pattern regarding FLAIR hyperintensities, mass effect, and contrast-enhancement around the surgical cavity (BT-RADS 2). - BT-RADS 3a: reflects worsening related to treatment effects, including radiation therapy and medications
Fig 7: BT-RADS 3a example. A and C are fluid-attenuated inversion recovery (FLAIR) images, while B and D are contrast-enhanced T1-weighted images from two consecutive magnetic resonance imaging (MRI) scans. Patient history includes resection of right frontal lobe glioblastoma in October 2023, followed by radiation therapy completed in January 2024. The patient additionally commenced treatment with temozolomide. The February 2024 MRI reveals an increase in FLAIR hyperintensity, mass effect, and contrast-enhancement along the surgical cavity margins, a worsening potentially related to radiation therapy effect, as less than 90 days have passed since the treatment (BT-RADS 3a). - BT-RADS 3b: reflects an indeterminate mix of treatment effect and tumor worsening
Fig 8: BT-RADS 3b example. A and C are fluid-attenuated inversion recovery (FLAIR) images, while B and D are contrast-enhanced T1-weighted images from two consecutive magnetic resonance imaging (MRI) scans of the same patient in Figure 10 and 11. Patient history includes resection of right frontal lobe glioblastoma in March 2014, followed by two radiotherapy treatments in June 2014 and May 2015. The November 2015 MRI reveals a substantially stable FLAIR hyperintensity and mass effect, and an increase in contrast-enhancement along the posterior side of the surgical cavity, a worsening potentially related to a mix of radiation therapy effect and tumor progression, as more than 90 days have passed since the last radiotherapy (BT-RADS 3b). - BT-RADS 3c: reflects increasing tumor burden
Fig 9: BT-RADS 3c example. A and C are fluid-attenuated inversion recovery (FLAIR) images, while B and D are contrast-enhanced T1-weighted images from two consecutive magnetic resonance imaging (MRI) scans. Patient history includes resection of left frontal lobe glioblastoma in September 2021, followed by radiation therapy completed in December 2021. The patient additionally commenced treatment with temozolomide. The January 2023 MRI demonstrates an increase in FLAIR hyperintensity within the surgical cavity and an increase in contrast-enhancement surrounding the surgical cavity, indicating a qualitative worsening quantified as less than 25% of the total brain lesion (BT-RADS 3c). - BT-RADS 4: reflects definite tumor progression
Fig 10: BT-RADS 4 example. A and C are fluid-attenuated inversion recovery (FLAIR) images, while B and D are contrast-enhanced T1-weighted images from two consecutive magnetic resonance imaging (MRI) scans of the same patient in Figure 8 and 11. Patient history includes resection of right frontal lobe glioblastoma in March 2014, followed by two radiotherapy treatments in June 2014 and May 2015. The February 2016 MRI demonstrates an increase in FLAIR hyperintensity and in contrast-enhancement surrounding the surgical cavity, indicating a qualitative worsening quantified as more than 25% of the total brain lesion (BT-RADS 4).
Fig 11: BT-RADS 4 example. A and C are fluid-attenuated inversion recovery (FLAIR) images, while B and D are contrast-enhanced T1-weighted images from two consecutive magnetic resonance imaging (MRI) scans of the same patient in Figure 8 and 10. Patient history includes resection of right frontal lobe glioblastoma in March 2014, followed by two radiotherapy treatments in June 2014 and May 2015. The patient additionally commenced treatment with fotemustine. The June 2016 MRI demonstrates a slight increase in FLAIR hyperintensity, mass effect, and in contrast-enhancement in the genu of the corpus callosum, indicating a qualitative worsening quantified as less than 25% of the total brain lesion. However, considering that a BT-RADS 4 had already been assigned in the previous examination (see Figure 10), indicative of progressive worsening, the final score will again be BT-RADS 4 and not BT-RADS 3c.
Fig 12: BT-RADS 4 example. A and C are fluid-attenuated inversion recovery (FLAIR) images, while B and D are contrast-enhanced T1-weighted images from two magnetic resonance imaging (MRI) scans of the same patient in Figure 4 and 5. Patient history includes resection of left parietal lobe recurrent glioblastoma in December 2015 and in September 2016, followed by radiation therapy completed in November 2016. The patient additionally commenced treatment with bevacizumab in November 2016. The September 2020 MRI demonstrates the appearance of a new lesion with contrast-enhancement in the left temporal pole, outside from the radiation treatment zone, definitively suspicious for tumor recurrence (BT-RADS 4).
Moreover, BT-RADS includes a structured reporting format with sections covering clinical indication, technique, comparative analyses, tumor-related and additional findings, and overall impression
To improve the usability and accessibility of the BT-RADS, an educational website was launched in May 2018 (https://btrads.com/). This platform offers public access to the scoring system's diagnostic flowchart and includes an interactive tool to guide healthcare professionals in applying the BT-RADS score to patient cases.
BT-RADS's prognostic and diagnostic value
Misclassifying high-grade gliomas as recurrent or non-recurrent can significantly impact patient care. Overestimating recurrence may lead to unnecessary invasive procedures and potentially harmful therapies, while underestimating recurrence may delay crucial interventions, ultimately compromising patient survival. Preliminary evidence from five studies indicates that BT-RADS may effectively identify tumor progression. Notably, category 3b appears to be a significant threshold, suggesting a high likelihood of recurrence and a poorer prognosis. Based on these findings, patients classified as 3b or higher may potentially benefit from a more proactive management approach.
BT-RADS's reliability
To ensure consistent and reliable application across various healthcare settings, the BT-RADS system, like other established radiological scoring systems, requires rigorous validation through inter-rater reliability studies. Low inter-reader agreement would hinder the standardization and widespread adoption of the system. Five initial studies have highlighted the robust reliability of BT-RADS. Moreover, the BT-RADS can function as an educational resource, assisting less experienced or junior radiologists in interpreting brain tumor imaging.
BT-RADS's structured report quality
Brain tumor management necessitates frequent, critical decisions often guided by imaging interpretation within a MDT setting. These decisions can significantly impact patient outcomes, influencing initial diagnoses, surgical approaches, and therapeutic plans. Therefore, it is crucial that the BT-RADS system be universally understood and adopted by all members of the MDT. Evidence from four preliminary studies indicates that the BT-RADS structured reporting format improves the exchange of information within MDTs treating brain tumors. This standardized framework ensures that all relevant information is effectively communicated among specialists, facilitating efficient and informed decision-making.
Future perspectives
While BT-RADS primarily utilizes FLAIR and contrast-enhanced T1-weighted images, ensuring widespread applicability, a comprehensive assessment may require the integration of additional imaging techniques. In complex scenarios, advanced modalities such as perfusion-weighted imaging (PWI) and diffusion-weighted imaging (DWI) could provide valuable insights, aiding in the discrimination between tumor recurrence and cerebral radiation necrosis. Moreover, the use of artificial intelligence could make it possible to semi-automatically segment tumor lesions and subsequently classify them according to the BT-RADS categories. The BT-RADS authors are currently collaborating with the American College of Radiology (ACR) to integrate the BT-RADS into the existing well-known RADS supported by the ACR. This collaborative effort involves incorporating findings from published research, which may lead to future updates and refinements of the BT-RADS scoring algorithm.
Table 2 summarizes the preliminary studies that explored the utility of BT-RADS in the surveillance of adult-type diffuse gliomas
