Skip to main content
Download PDF

Comparison of a new MR rapid wash-out map with MR perfusion in brain tumors

BMC Cancer volume 24, Article number: 1139 (2024) Cite this article

Abstract

Background

MR perfusion is a standard marker to distinguish progression and therapy-associated changes after surgery and radiochemotherapy for glioblastoma. TRAMs (Treatment Response Assessment Maps) were introduced, which are intended to facilitate the differentiation of vital tumor cells and radiation necrosis by means of late (20–90 min) contrast clearance and enhancement. The differences of MR perfusion and late-enhancement are not fully understood yet.

Methods

We have implemented and established a fully automated creation of rapid wash-out (15–20 min interval) maps in our clinic. We included patients with glioblastoma, CNS lymphoma or brain metastases who underwent our MR protocol with MR perfusion and rapid wash-out between 01/01/2024 and 30/06/2024. Since both wash-out and hyperperfusion are intended to depict the active tumor area, this study involves a quantitative and qualitative comparison of both methods. For this purpose, we volumetrically measured rCBV (relative cerebral blood volume) maps and rapid wash-out maps separately (two raters). Additionally, we rated the agreement between both maps on a Likert scale (0–10).

Results

Thirty-two patients were included in the study: 15 with glioblastoma, 7 with CNS lymphomas and 10 with brain metastasis. We calculated 36 rapid wash-out maps (9 initial diagnosis, 27 follow-up).

Visual agreement of MR perfusion with rapid wash-out by rating were found in 44 ± 40% for initial diagnosis, and 75 ± 31% for follow-up. We found a strong correlation (Pearson coefficient 0.92, p < 0.001) between the measured volumes of MR perfusion and rapid wash-out. The measured volumes of MR perfusion and rapid wash-out did not differ significantly. Small lesions were often not detected by MR perfusion. Nevertheless, the measured volumes showed no significant differences in this small cohort.

Conclusions

Rapid wash-out calculation is a simple tool that provides new information and, when used in conjunction with MR perfusion, may increase diagnostic accuracy. The method shows promising results, particularly in the evaluation of small lesions.

Peer Review reports

Introduction

The classic analysis of contrast dynamics in brain tumors in MRI is limited to contrast enhancement and contrast agent flooding using MR perfusion. MR perfusion is the current standard for the differentiation of tumor recurrence and treatment related changes in the follow-up of patients with high grade gliomas [1, 2]. It also plays a crucial role in the initial diagnosis, especially in distinguishing gliomas from CNS lymphoma [3].

The question of what happens to the contrast agent in the tumor and its surroundings at a later point in time has long been neglected in diagnosis finding. The basic concept of a late contrast clearance (> 60 min) for brain tumors was introduced by Zach et al. in 2012 [4].

It has been shown that this method called TRAMs (Treatment Response Assessment Maps) works very well in distinguishing tumor progression from radiation necrosis or pseudoprogression in glioblastoma [5], and helps to assess the response to therapy in CNS lymphomas [6] or metastasis [7]. A study from Satvat et al. [8] revealed that even a shorter waiting time (< 20 min) between the two T1-weighted sequences can produce meaningful results.

Firstly, we noticed a discrepancy between the protocols of both methods from Zach et al. and Satvat et al. and the current literature. In the initial protocols [4, 8], the first contrast-enhanced, T1-weighted sequence starts directly / 2 min after MR perfusion. But, the maximum Gadolinium enhancement of glioblastoma cells is described at approx. 3–8 min [9, 10] in rat models, and in metastasis even at approx. 10–15 min [11] after contrast agent applications.

For this reason, we initially incorporated a delay sequence into our MR protocol, as we wanted to ensure that the maximum enhancement of glioblastoma cells had already been achieved in the first T1-weighted sequence.

Secondly, we optimized our calculated rapid wash-out maps for the semi-automated volumetric analysis with 3D slicer and normalized the maps for future processing using neural networks. We used a simple two-color-system with red for wash-out (active tumor cells / vessel) and green for wash-in (scar tissue / necrosis). Figure 1 demonstrates an example of the rapid wash-out maps and comparison with MR perfusion.

Fig. 1
figure 1

left: transverse rapid wash-out map, wash-out (red) and wash-in (green); right: transverse MR perfusion at the same level; two different heights of the tumor are cut above and below; A – mismatch between MR perfusion and rapid wash-out; B - partially matching tumor area; C – tumor area active in both MR perfusion and wash-out

Full size image

In this first part of our study, we wanted to demonstrate the feasibility and compare our rapid wash-out mapping with MR perfusion for glioblastoma, CNS lymphomas and brain metastases, both at initial diagnosis and while treatment.

Methods

Study design

This single-center observational study was ethically approved by the institutional review board. This study adhered to the 2013 Declaration of Helsinki. The institutional review board waived the requirement for informed consent because of the retrospective nature of the study and the analysis of anonymized patient data. All methods were performed in accordance with relevant guidelines and regulations.

Participant population

We searched our PACS database for patients with valid rapid wash-out protocols between 01/01/2024 and 30/06/2024. We included only patients with a diagnosis of glioblastoma, CNS lymphoma or brain metastasis, and divided the found protocols into initial diagnosis and follow-up MRI.

MRI protocol and technic details

We performed protocols on Siemens Sola 1.5T (Siemens Healthineers AG, Werner-von-Siemens-Str. 1, D-80333 Munich, Germany), Philips Achieva 3T and Philips Intera 1.5T (Koninklijke Philips NV, Amstelplein 2, Amsterdam, 1096 BC, Netherlands).

The detailed protocols are listed in Table 1. We acquired at three times the same T1-weighted sequence (native, 5 min and 25 min after contrast application): sagittal T1 MPRAGE on Siemens and transversal T1w 3D TFE on Philips MRI. If no contrast enhancement in the first contrast-enhanced T1-weighted sequence was detected, a neuroradiologist shortened the protocol accordingly to optimize clinical workflow.

Table 1 Detailed MR protocols
Full size table

See Fig. 2 for a schematic overview of our protocol with estimated contrast enhancement curves.

Fig. 2
figure 2

Time line of MRI protocol (bottom; exemplary for 1.5T Sola Siemens) and schematic overview of estimated contrast enhancement curves (top)

Full size image

Sequence parameters for T1-weighted imaging were TE = 3.02 ms, TR = 1680 ms, flip angle = 8° for sagittal T1 MPRAGE (slice-thickness 1 mm, resolution 1 × 1 mm, scan time 342 s, 1.5T Sola, Siemens); TE = 3.5 ms, TR = 7.9 ms, flip angle = 8° for T1w 3D TFE (slice-thickness 1 mm, resolution 1 × 1 mm, scan time 258 s, 3T Achieva, Philips); and TE = 5.9 ms, TR = 24 ms, flip angle = 25° for transversal T1w 3D TFE (slice-thickness 1 mm, resolution 1 × 1 mm, scan time 280 s, 1.5T Intera, Philips).

Perfusion protocols and evaluation

Single dose MR perfusion was performed using manufacturer dependant DSC (Dynamic Susceptibility Contrast) T2* sequences (Siemens EP2D_Perfusion, Philips FFE_EPI_HR_Perfusion) without preload, and a bolus application of 0,1 ml Gadovist (Gadobust, Bayer Vital GmbH, Gebäude K56, D-51366 Leverkusen, Germany) per kg body weight with a flow of 3 ml / second. The perfusion curves were evaluated and rCBV was measured and calculated in the grey scale CBV maps as described by Cho et al. [12] with T2* MR Neuroperfusion Tool of Philips Intellispace or MR Neurology Perfusion Tool of Siemens syngo.via. We exported the rCBV maps to our PACS system for further analysis.

Volumetric measurements were performed for all tumor regions with contrast enhancement and elevated rCBV. Since rCBV-values of tumor regions can be low, especially in CNS lymphomas [13, 14], we choose a significant and optically distinguishable threshold of rCBV > 1.2. We did this to define as large a volume as possible using MR perfusion in this first feasibility study, although we knew that the optimal CBV threshold for the differentiation of pseudoprogression and recurrent high-grade gliomas seems to be between > 1.64 and > 1.75 [15]. We used the segment editor of the 3D slicer tool (www.slicer.org) to measure these volumes.

Sequence parameters for transversal T2* perfusion weighted imaging were TE = 31 ms, TR = 2840 ms, flip angle = 90° for EP2D_Perfusion (slice-thickness 4 mm, resolution 1.8 × 1.8 mm, scan time 151 s, 1.5T Sola, Siemens); TE = 40 ms, TR = 1552 ms, flip angle = 75° for FFE_EPI_HR_Perfusion (slice-thickness 4 mm, resolution 1.75 × 1.75 mm, scan time 67 s, 3T Achieva, Philips); and TE = 40 ms, TR = 1952 ms, flip angle = 75° for FFE_EPI_HR_Perfusion (slice-thickness 4 mm, resolution 2.33 × 2.33 mm, scan time 86 s, 1.5T Intera, Philips).

Calculation of wash-out-maps

The wash-out masks are generated using Python 3.11 software developed in-house. First, both MRI series are converted to NIFTI format using the dicom2nifti library (version 2.4.10). The N4 bias field correction is applied to these images with a 3-fold reduction using SimpleITK (version 2.3.1). We then normalized both series to the same mean value and rigidly registered using the FLIRT class of the FSL library (version 6.0). Finally, the registered images are subtracted (series 1 minus series 2) and converted into both gray and RGB DICOM image series. For the RGB series, the used value range is first halved and then normalized in the middle to 127 out of 255. Negative subtraction values are colored green and positive values are colored red. Thus, higher values in the first MRI appear red while lower values appear green.

Semi-automatically volumetric assessment

We measured MR perfusion, wash-out and wash-in volumes for every patient via 3D slicer (level trace tool, slice by slice, semi-automatically). Figure 3 demonstrates an exemplary measurement of the volumes. The measurement were performed independently by to raters and finally averaged. ICC(2, k) was calculated. We also calculated the wash-out ratio as wash-out-volume / (wash-out-volume + wash-in-volume), previously described in a TRAMs study [16].

Fig. 3
figure 3

Screenshot of the 3D slicer with an example segmentation of a brain metastasis of a malignant melanoma three months after radiotherapy

Full size image

Visual assessment and comparison

To avoid measuring unassociated volumes, the following additional ratings were used: We visually compared volumes segmented in MR perfusion and rapid wash-out maps slice per slice and rated the agreement of both on a 11-point-Likert scale 0–10 (0-100% agreement). Additionally, the following estimations were rated on a 11-point-Likert scale 0–10: (1) MR hyperperfusion not included in wash-out compared with whole hyperperfusion volume, and (2) wash-out not included in MR hyperperfusion compared with whole wash-out volume.

Statistical analysis

We used Python3 (Version 3.11, matPlotLib) for statistical programming and histogram / image construction. We calculated the intraclass correlation coefficient (ICC, 2, k) using Python3 and the libraries “pandas” and “pingouin”. Interpretation of results was done following Koo and Li [17]. We used the Pearson coefficient for correlation analysis and Tukey’s Test for analysis of variance [18]. The significance level was set to 5%.

Results

Participants

Our PACS search revealed 51 rapid wash-out protocols for 45 patients. We excluded 11 patients (13 protocols) without diagnosis of glioblastoma, CNS lymphoma or brain metastasis. At evaluation, we had to exclude two patients (2 protocols) because of severe movement artifacts. See Fig. 4 for a detailed flow chart. We formed six subgroups: initial diagnosis of (1) glioblastoma, (2) CNS lymphoma, (3) brain metastases; and follow-up MRI of (4) glioblastoma, (5) CNS lymphoma, and (6) brain metastases.

Fig. 4
figure 4

Inclusion flow chart

Full size image

We finally included 32 patients (18 females, 14 males). Mean patients age was 67 ± 11 (mean ± standard deviation) years. Thirty-six rapid wash-out maps were included (11 on 1.5T Sola, 24 on 3.0T Achieva and one on 1.5T Intera).

Calculation algorithm and time

The running time of the algorithms was mainly dependent on the registration task of the FLIRT class, which took between one and eight minutes. The remaining running time was less than one minute, so that all calculations took less than 10 min each. Thus, the calculation can be performed on a normal PC in almost real time.

Histology

All included patients with high-grade glioma had a diagnosis of IDH-wildtype glioblastoma (WHO grade 4). The seven included CNS lymphoma patients suffered from a non-Hodgkin B-cell lymphoma. In the patients with initial diagnosis of brain metastases, we found the following histopathologies: adenocarcinoma of the lung, small cell lung cancer, transitional cell carcinoma and esophageal squamous-cell carcinoma. Pathologies for patient with brain metastases in follow-up scans were malignant melanoma, squamous cell carcinoma, small cell lung cancer and esophageal squamous-cell carcinoma.

Visual comparison

We visually rated the agreement of MR perfusion and rapid wash-out per lesion. Table 2 demonstrates the results of this subjective evaluation. Since this rating was given per lesion and especially small lesions were often not detected by MR perfusion, large statistical differences were observed (especially for CNS lymphomas having small lesions in our cohort). While the MR perfusion was mostly almost completely contained in the wash-out regions, we also found large areas of wash-out that were not shown in the MR perfusion. Figure 5 reveals typical initial diagnosis imaging of DSC perfusion and rapid wash-out. During follow-up, the differences were not so notable, but this was also due to the fact that the CNS lymphomas in particular mostly showed a good response. Follow-up of glioblastoma are shown in Fig. 6.

Fig. 5
figure 5

Examples of initial diagnosis (top: glioblastoma; middle: CNS lymphoma; bottom: brain metastasis of a small cell lung carcinoma). Left: contrast enhanced T1-weighted images; middle rCBV map of MR perfusion; right: rapid wash-out map

Full size image
Fig. 6
figure 6

Glioblastoma therapy assessment. Top: 200% volume progress in 6-week follow-up. MR perfusion (middle) showed no local hyperperfusion. Rapid wash-out (right) revealed areas of increased rapid wash-out (red). Middle: Typical post-therapeutically changes without hyperperfusion and wash-out, but wash-in (green). Bottom: Radiation necrosis (3-month follow-up was stable) with less active tumor cells with bubble-like appearance of small contrast enhancing lesions and minimal hyperperfusion (arrow) as well as minimal ring-shaped wash-outs with red dot (arrow). Left: contrast enhanced T1-weighted images; middle rCBV map of MR perfusion; right: rapid wash-out map

Full size image
Table 2 Mean ± standard deviation of the patient data and agreement of volumes on 11-point likert scale (0-100%)
Full size table

Volumetric compare

Due to the different voxel sizes and the possible partial volume effect, the data should be viewed with caution. The results of volumetric measurement of MR perfusion and rapid wash-out maps are listed in Table 3.

Table 3 Mean ± standard deviation of the volumetric results
Full size table

We found a strong correlation between the measured volumes of MR perfusion and rapid wash-out maps with a Pearson correlation coefficient r of 0.92, (p < 0.001; R²=0.84).For wash-out ratios at initial diagnosis, the number of lesions was not sufficient to make a statistical statement. Wash-out ratios of glioblastomas on follow-up were significantly higher than those of CNS lymphoma (t-Test, p < 0.001), and brain metastasis (t-Test, p < 0.01).

The volumes that could be measured in MR perfusion compared with the correlating rapid wash-out volumes revealed no significant differences (p > 0.5 for each t-Test “perfusion versus wash-out”) for five subgroups (initial diagnosis: GBM, CNS lymphoma, brain metastasis; and post-therapeutic: GBM and brain metastasis). The only exception were the post-therapeutic CNS lymphomas (p < 0.03) where no volumes could be measured in MR perfusion due to the minimal size of the lesions. The additionally calculated analysis of variance of the six subgroups with Tukey’s Test revealed no significant models / results, mainly caused by the low number of cases.

Intraclass correlation

Intraclass correlation was excellent with ICC (2, k) = 0.979, confidence interval [0.95, 0.99] for all measured volumes (wash-in, wash-out and MR perfusion).

Discussion

In our feasibility study of 32 patients, we were able to demonstrate the regular, standardized and meaningful use of our new rapid wash-out maps. Even with a short delay of 15 min, we were able to generate useful maps on which the wash-out and wash-in could be easily understood. Semi-automatically volumetrization could be performed easily. We were able to show similarities with MR perfusion, but also revealed differences and potential for additional information.

The new rapid wash-out maps can be volumetrically assessed, are calculated in nearly real-time (10 min), and are optimized for further post-processing, e.g. using AI-tools or neuronal nets.

Especially in cases of radiation necrosis, pseudo-progression and tumor progress, we already found individual cases demonstrating diagnostic differences between wash-out maps and MR perfusion (see Fig. 6).

The new method is also well usable to CNS lymphomas and metastases, although the results, especially in metastases, seem to depend on the histology and need to be investigated in larger studies.

The evaluation of the volumetry did not provide any statistically significant information, since the areas of hyperperfusion and wash-out often differed partially in terms of their location, but differed only slightly in terms of their overall size.

Following RANO (Response Assessment in Neuro-Oncology) criteria [19], and RANO 2.0 [20], for high-grade glioma, in patients with high likelihood of pseudoprogression, mandatory confirmation of progression with a repeat MRI is recommended, volumetric measurements also are an option. This currently applies to therapies with the STUPP scheme [21]. Advance imaging methods, such as ADC (apparent diffusion coefficient)-maps and MR perfusion, are generally recommended. Contrast clearance or late-enhancement is not specifically addressed in these guidelines. RANO for brain metastasis [22] and iRANO (immunotherapy Response Assessment in Neuro-Oncology) [23] for immunotherapy (malignant melanoma, lung cancer, glioblastoma) refine the diagnostic process.

Radiation necrosis and pseudo-progression occur in 10 to 35% of all cases, depending on the study, and represent the main problem in diagnosis [24].

MR perfusion is the most used method to differ progressive disease from radiation necrosis and pseudo-progression [2]. However, there is currently no reliable method of distinguishing between the two [25].

In glioblastoma, the measurement of MR perfusion parameters showed promising results, but seem to vary from study to study, e.g. a study from UK found a high rCBV-threshold of 3.0 [26]. A meta-analysis revealed the following pooled sensitivities and specificities for the differentiation of glioma recurrence and pseudo progression in MR perfusion techniques: 0.82, 0.87 for DSC, 0.83, 0.83 for DCE (Dynamic Contrast Enhanced), and 0.78, 0.86 for ASL (Arterial Spin Labeling) perfusion [27]. Sensitivity and specificity of TRAMs are approximately 0.93 and 0.78 [28].

A combination of ADC and DSC perfusion showed no significant improvement of the diagnostic accuracy compared to each method alone [29]. Post-operative DWI (diffusion-weighted imaging) can be improved by other diffusion weighted imaging methods like STEAM (Stimulated Echo Acquisition Mode) DWI [30], which may increase the diagnostic accuracy after surgery and therefore also in the 3-months follow-up.

A comparative study of MR perfusion versus contrast clearance/TRAMs versus (18 F)-dopa PET/CT showed different advantages and disadvantages of the individual methods, in favor of the two MR methods [7]. Another alternative, MR spectroscopy demonstrated excellent results in some studies [31], but is still matter of debate, because it seems to be relatively susceptible to artifacts and is examiner-dependent.

In immunotherapy for glioblastoma, TRAMs may be useful as well [32]. Some studies also demonstrated that TRAMs can be used in treatment assessment of CNS lymphomas [6, 33].

Treatment-assessment in brain metastasis is also possible after stereotactic surgery [34], but with a larger overlap between tumor recurrence and radiation necrosis. This is possibly due to the delay being too short, so that the medium uptake in metastasis had not yet reached the maximum wash-in and then finally showed the same value as slow-uptake necrosis. Other studies after gamma-knife surgery revealed improved results [35] comparable with MR perfusion [36].

Limitations

This first study is only intended as a feasibility study. The small number of cases does allow only a few statistically significant statements.

Main limitations of the study were the small sample size, and the partially old protocols and MR scanners. Especially the MR DSC perfusion protocols deviate in some cases considerably from the current consensus recommendations [37].

Additionally, on Siemens MR scanner, we used T1 MPRAGE sequences, optimized for grey-white-matter contrast and not for contrast enhancement. This could reduce the conspicuity of small lesions and reduce the contouring estimates compared with T1 VIBE sequences, as described by Danieli et al. [38].

The short time difference between the contrast-enhanced T1-weighted sequences compared to the TRAMs could reduce the wash-out volume.

In particular, the wash-in/late-enhancement behavior of the different metastasis subtypes is not clear (with low wash-out ratios), but this may also open up new diagnostic perspectives.

In addition, the delay before the first T1-weighted sequence is optimized for glioblastomas. In certain circumstances, a different delay is recommended depending on the histology (possibly also depending on the primary tumor in the case of brain metastases). There is still a big research gap that needs to be filled.

Outlook

The rapid wash-out is intended as a follow-up test and not just to record the current status. The following studies must show to what extent these rapid wash-out maps can provide useful new information for therapy assessment. It is not expected that these are better than TRAMs. However, they are much easier to integrate into everyday clinical practice, and the question is whether there is any inferiority of our rapid wash-out maps versus TRAMs. We do not expect this to be the case in glioblastomas and CNS lymphomas in particular, and in metastases it probably depends on the subtype. Furthermore, we think that the TRAMs could also be improved by optimizing the timing of the first T1-weighted sequence dependent on the histology. The exact advantages and disadvantages of rapid wash-out-maps compared to MR perfusion and contrast clearance analysis must be investigated in further prospective studies with specific questions.

In future analysis the combination of methods, e.g. DSC perfusion and FDG-PET (Fludeoxyglucose-18 Positron Emission Tomography) for high-grade gliomas [39]; or DTI (Diffusion Tensor Imaging) and DSC perfusion [40], will increase diagnostic safety.

Additional information with possible improvement of diagnostic accuracy could also be achieved by new PET tracers [41,42,43].

Even in difficult diagnostic tasks, such as the differentiation of multiple space-occupying lesions, the wash-out can provide additional information [44]. A meta-analysis revealed, that combined MRI techniques also enable a better diagnostic accuracy for the treatment response in brain metastasis [45]. However, more prospective and larger studies are needed [46].

The targeted use of AI tools [47] or Machine learning models [48] could also contribute to an improvement as well as the use of new MRI techniques, e.g. T1 mapping [49] for the assessment of standardized values.

Conclusion

We have successfully tested a new rapid wash out with a delayed first contrast-enhanced T1-weighted sequence. Our results reveal that rapid wash-out maps appear to be at least an equally good marker compared to MR perfusion. Especially the combination of tumor volume and wash-out ratio appears to be a promising method for monitoring disease progression.

Availability of data and materials

The datasets used and/or analyzed during the current study available from the corresponding author on reasonable request.

Abbreviations

ADC:

Apparent Diffusion Coefficient

AI:

Artificial Intelligence

ASL:

Arterial Spin Labeling (MR perfusion)

BM:

Brain Metastases

CCA:

Contrast Clearance Analysis

DCE:

Dynamic Contrast Enhanced (MR perfusion)

DSC:

Dynamic Susceptibility Contrast (MR perfusion)

DTI:

Diffusion Tensor Imaging

DWI:

Diffusion-Weighted Imaging

FDG-PET:

Fludeoxyglucose-18 Positron Emission Tomography

Gb:

Glioblastoma

iRANO:

immunotherapy Response Assessment in Neuro-Oncology

MPRAGE:

Magnetization Prepared with RApid Gradient Echo

MRI:

Magnetic Resonance Imaging

MR:

Magnetic Resonance

rCBV:

relative Cerebral Blood Volume

RANO:

Response Assessment in Neuro-Oncology

SD:

Standard Deviation

STEAM:

Stimulated Echo Acquisition Mode

TRAMs:

Treatment Response Assessment Maps

VIBE:

Volumetric Interpolated Breath-hold Examination

WHO:

World Health Organization

References

  1. Patel P, Baradaran H, Delgado D, Askin G, Christos P, John Tsiouris A, et al. MR perfusion-weighted imaging in the evaluation of high-grade gliomas after treatment: a systematic review and meta-analysis. Neuro-Oncol. 2017;19:118–27.

    Article  PubMed  Google Scholar 

  2. Thust SC, Heiland S, Falini A, Jäger HR, Waldman AD, Sundgren PC, et al. Glioma imaging in Europe: a survey of 220 centres and recommendations for best clinical practice. Eur Radiol. 2018;28:3306–17.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  3. Abreu VS, Tarrio J, Silva J, Almeida F, Pinto C, Freitas D, et al. Multiparametric analysis from dynamic susceptibility contrast-enhanced perfusion MRI to evaluate malignant brain tumors. J Neuroimaging. 2024;34:257–66.

    Article  PubMed  Google Scholar 

  4. Zach L, Guez D, Last D, Daniels D, Grober Y, Nissim O, et al. Delayed contrast Extravasation MRI for depicting tumor and Non-tumoral Tissues in primary and metastatic brain tumors. PLoS ONE. 2012;7:e52008.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  5. Zach L, Guez D, Last D, Daniels D, Grober Y, Nissim O, et al. Delayed contrast extravasation MRI: a new paradigm in neuro-oncology. Neuro-Oncol. 2015;17:457–65.

    Article  CAS  PubMed  Google Scholar 

  6. Millard T, Chau I, Iyengar S, El-Sharkawi D, Cunningham D, Sharma B. Treatment Response Assessment maps (TRAMs), a new tool for CNS lymphoma. eJHaem. 2022;3:247–8.

    Article  PubMed  PubMed Central  Google Scholar 

  7. Lakehayli Z, Phlips P, Margoum A, Saoudi A, Hmaid L, Nejjar I, et al. What effective technique to differentiate radiation brain necrosis from a tumor progression in patients treated with radiation: a monocentric retrospective study combining the MRI TRAMs technique and the (18F)-dopa PET/CT. Cancer/Radiothérapie. 2023;27:273–80.

    Article  CAS  PubMed  Google Scholar 

  8. Satvat N, Korczynski O, Müller-Eschner M, Othman AE, Schöffling V, Keric N, et al. A Rapid Late Enhancement MRI Protocol improves differentiation between Brain Tumor recurrence and treatment-related contrast Enhancement of Brain Parenchyma. Cancers. 2022;14:5523.

    Article  PubMed  PubMed Central  Google Scholar 

  9. Attenberger UI, Runge VM, Jackson CB, Baumann S, Birkemeier K, Michaely HJ, et al. Comparative evaluation of Lesion Enhancement using 1 M gadobutrol vs. 2 conventional gadolinium chelates, all at a dose of 0.1 mmol/kg, in a rat brain tumor model at 3 T. Invest Radiol. 2009;44:251–6.

    Article  CAS  PubMed  Google Scholar 

  10. Engelhorn T, Schwarz MA, Eyupoglu IY, Kloska SP, Struffert T, Doerfler A. Dynamic contrast enhancement of experimental glioma. Acad Radiol. 2010;17:188–93.

    Article  PubMed  Google Scholar 

  11. Kushnirsky M, Nguyen V, Katz JS, Steinklein J, Rosen L, Warshall C, et al. Time-delayed contrast-enhanced MRI improves detection of brain metastases and apparent treatment volumes. J Neurosurg. 2016;124:489–95.

    Article  CAS  PubMed  Google Scholar 

  12. Cho SK, Na DG, Ryoo JW, Roh HG, Moon CH, Byun HS, et al. Perfusion MR Imaging: clinical utility for the Differential diagnosis of various brain tumors. Korean J Radiol. 2002;3:171.

    Article  PubMed  PubMed Central  Google Scholar 

  13. Toh CH, Wei K-C, Chang C-N, Ng S-H, Wong H-F. Differentiation of primary Central Nervous System Lymphomas and Glioblastomas: comparisons of diagnostic performance of dynamic susceptibility contrast-enhanced perfusion MR Imaging without and with contrast-Leakage correction. Am J Neuroradiol. 2013;34:1145–9.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  14. Lee MD, Baird GL, Bell LC, Quarles CC, Boxerman JL. Utility of Percentage Signal Recovery and Baseline Signal in DSC-MRI Optimized for Relative CBV Measurement for Differentiating Glioblastoma, Lymphoma, Metastasis, and Meningioma. AJNR Am J Neuroradiol. 2019;40(9):1445–50. https://doi.org/10.3174/ajnr.A6153.

  15. Anil A, Stokes AM, Chao R, Hu LS, Alhilali L, Karis JP, et al. Identification of single-dose, dual-echo based CBV threshold for fractional tumor burden mapping in recurrent glioblastoma. Front Oncol. 2023;13:1046629.

    Article  PubMed  PubMed Central  Google Scholar 

  16. Müller SJ, Khadhraoui E, Ganslandt O, Henkes H, Gihr GA. MRI Treatment Response Assessment maps (TRAMs) for differentiating recurrent glioblastoma from radiation necrosis. J Neurooncol. 2024. https://doi.org/10.1007/s11060-024-04573-x.

    Article  PubMed  Google Scholar 

  17. Koo TK, Li MY. A Guideline of selecting and reporting Intraclass correlation coefficients for Reliability Research. J Chiropr Med. 2016;15:155–63.

    Article  PubMed  PubMed Central  Google Scholar 

  18. Tukey JW. Comparing Individual Means in the analysis of Variance. Biometrics. 1949;5:99.

    Article  CAS  PubMed  Google Scholar 

  19. Wen PY, Macdonald DR, Reardon DA, Cloughesy TF, Sorensen AG, Galanis E, et al. Updated response Assessment Criteria for High-Grade gliomas: Response Assessment in Neuro-Oncology Working Group. J Clin Oncol. 2010;28:1963–72.

    Article  PubMed  Google Scholar 

  20. Wen PY, Van Den Bent M, Youssef G, Cloughesy TF, Ellingson BM, Weller M, et al. RANO 2.0: update to the Response Assessment in Neuro-Oncology Criteria for High- and low-Grade gliomas in adults. J Clin Oncol. 2023;41:5187–99.

    Article  PubMed  Google Scholar 

  21. Stupp R, Mason WP, Van Den Bent MJ, Weller M, Fisher B, Taphoorn MJB, et al. Radiotherapy plus Concomitant and Adjuvant Temozolomide for Glioblastoma. N Engl J Med. 2005;352:987–96.

    Article  CAS  PubMed  Google Scholar 

  22. Lin NU, Lee EQ, Aoyama H, Barani IJ, Barboriak DP, Baumert BG, et al. Response assessment criteria for brain metastases: proposal from the RANO group. Lancet Oncol. 2015;16:e270–8.

    Article  PubMed  Google Scholar 

  23. Okada H, Weller M, Huang R, Finocchiaro G, Gilbert MR, Wick W, et al. Immunotherapy response assessment in neuro-oncology: a report of the RANO working group. Lancet Oncol. 2015;16:e534–42.

    Article  PubMed  PubMed Central  Google Scholar 

  24. Zikou A, Sioka C, Alexiou GA, Fotopoulos A, Voulgaris S, Argyropoulou MI. Radiation Necrosis, Pseudoprogression, Pseudoresponse, and Tumor recurrence: Imaging challenges for the Evaluation of Treated Gliomas. Contrast Media Mol Imaging. 2018;2018:1–6.

    Article  Google Scholar 

  25. Alexiou GA, Tsiouris S, Kyritsis AP, Voulgaris S, Argyropoulou MI, Fotopoulos AD. Glioma recurrence versus radiation necrosis: accuracy of current imaging modalities. J Neurooncol. 2009;95:1–11.

    Article  PubMed  Google Scholar 

  26. Maiter A, Butteriss D, English P, Lewis J, Hassani A, Bhatnagar P. Assessing the diagnostic accuracy and interobserver agreement of MRI perfusion in differentiating disease progression and pseudoprogression following treatment for glioblastoma in a tertiary UK centre. Clin Radiol. 2022;77:e568–75.

    Article  CAS  PubMed  Google Scholar 

  27. Zhang J, Wang Y, Wang Y, Xiao H, Chen X, Lei Y, et al. Perfusion magnetic resonance imaging in the differentiation between glioma recurrence and pseudoprogression: a systematic review, meta-analysis and meta-regression. Quant Imaging Med Surg. 2022;12:4805–22.

    Article  PubMed  PubMed Central  Google Scholar 

  28. Bodensohn R, Forbrig R, Quach S, Reis J, Boulesteix A-L, Mansmann U, et al. MRI-based contrast clearance analysis shows high differentiation accuracy between radiation-induced reactions and progressive disease after cranial radiotherapy. ESMO Open. 2022;7:100424.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  29. Prager AJ, Martinez N, Beal K, Omuro A, Zhang Z, Young RJ. Diffusion and perfusion MRI to Differentiate Treatment-related changes including pseudoprogression from recurrent tumors in high-Grade Gliomas with histopathologic evidence. Am J Neuroradiol. 2015;36:877–85.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  30. Müller SJ, Khadhraoui E, Voit D, Riedel CH, Frahm J, Romero JM, et al. Comparison of EPI DWI and STEAM DWI in early postoperative MRI controls after resection of tumors of the Central Nervous System. Clin Neuroradiol. 2023. https://doi.org/10.1007/s00062-023-01261-7.

    Article  PubMed  PubMed Central  Google Scholar 

  31. El-Abtah ME, Talati P, Fu M, Chun B, Clark P, Peters A, et al. Magnetic resonance spectroscopy outperforms perfusion in distinguishing between pseudoprogression and disease progression in patients with glioblastoma. Neuro-Oncol Adv. 2022;4:vdac128.

    Article  Google Scholar 

  32. Cuccarini V, Savoldi F, Mardor Y, Last D, Pellegatta S, Mazzi F, et al. Response assessment of GBM during immunotherapy by delayed contrast treatment response assessment maps. Front Neurol. 2024;15:1374737.

    Article  PubMed  PubMed Central  Google Scholar 

  33. Kowa JY, Millard T, Goldman A, Sharma RK, Attygalle A, Mahalingam P, Marshall K, Welsh L, Li S, Mackinnon A, Rich P, Nicholson E, Iyengar S, El-Sharkawi D, Chau I, Cunningham D, Sharma B. Are treatment response assessment maps (TRAMs) and 18 F-choline positron emission tomography the future of central nervous system lymphoma imaging? Br J Haematol. 2021;195(1):e116–e19. https://doi.org/10.1111/bjh.17632.

  34. Admojo L, Korte J, Anderson N, Phillips C, Caspersz L, Lasocki A. Investigating the role of delayed contrast magnetic resonance imaging (MRI) to differentiate radiation necrosis from tumour recurrence in brain metastases after stereotactic radiosurgery. J Med Imaging Radiat Oncol. 2023;67:292–8.

    Article  PubMed  Google Scholar 

  35. Goulenko V, Madhugiri VS, Bregy A, Recker M, Lipinski L, Fabiano A, et al. Histopathological correlation of brain tumor recurrence vs. radiation effect post-radiosurgery as detected by MRI contrast clearance analysis: a validation study. J Neurooncol. 2024;168:547–53.

    Article  PubMed  Google Scholar 

  36. Peker S, Samanci Y, Aygun MS, Yavuz F, Erden ME, Nokay AE, et al. The Use of Treatment Response Assessment maps in discriminating between Radiation Effect and Persistent Tumoral Lesion in metastatic brain tumors treated with Gamma Knife Radiosurgery. World Neurosurg. 2021;146:e1134–46.

    Article  PubMed  Google Scholar 

  37. Boxerman JL, Quarles CC, Hu LS, Erickson BJ, Gerstner ER, Smits M, et al. Consensus recommendations for a dynamic susceptibility contrast MRI protocol for use in high-grade gliomas. Neuro-Oncol. 2020;22:1262–75.

    Article  PubMed  PubMed Central  Google Scholar 

  38. Danieli L, Riccitelli GC, Distefano D, Prodi E, Ventura E, Cianfoni A, et al. Brain tumor-enhancement visualization and morphometric Assessment: a comparison of MPRAGE, SPACE, and VIBE MRI techniques. Am J Neuroradiol. 2019;40:1140–8.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  39. Panholzer J, Malsiner-Walli G, Grün B, Kalev O, Sonnberger M, Pichler R. Multiparametric analysis combining DSC-MR perfusion and [18F]FET-PET is Superior to a single parameter Approach for differentiation of Progressive Glioma from Radiation Necrosis. Clin Neuroradiol. 2024;34:351–60.

    Article  PubMed  Google Scholar 

  40. Wang S, Martinez-Lage M, Sakai Y, Chawla S, Kim SG, Alonso-Basanta M, et al. Differentiating Tumor Progression from Pseudoprogression in patients with Glioblastomas using Diffusion Tensor Imaging and Dynamic susceptibility contrast MRI. Am J Neuroradiol. 2016;37:28–36.

    Article  PubMed  PubMed Central  Google Scholar 

  41. De Marco R, Pesaresi A, Bianconi A, Zotta M, Deandreis D, Morana G, et al. A systematic review of amino acid PET imaging in adult-type high-Grade glioma surgery: a neurosurgeon’s perspective. Cancers. 2022;15:90.

    Article  PubMed  PubMed Central  Google Scholar 

  42. Ghosh S, Fletcher NL, Huda P, Houston ZH, Howard CB, Lund ME, et al. Pharmacokinetics and Biodistribution of 89 Zr-Miltuximab and its antibody fragments as Glypican-1 Targeting Immuno-PET agents in Glioblastoma. Mol Pharm. 2023;20:1549–63.

    Article  CAS  PubMed  Google Scholar 

  43. Santo G, Laudicella R, Linguanti F, Nappi AG, Abenavoli E, Vergura V, et al. The utility of conventional amino acid PET radiotracers in the Evaluation of Glioma Recurrence also in comparison with MRI. Diagnostics. 2022;12:844.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  44. Müller SJ, Khadhraoui E, Ernst M, Rohde V, Schatlo B, Malinova V. Differentiation of multiple brain metastases and glioblastoma with multiple foci using MRI criteria. BMC Med Imaging. 2024;24:3.

    Article  PubMed  PubMed Central  Google Scholar 

  45. Teunissen WHT, Govaerts CW, Kramer MCA, Labrecque JA, Smits M, Dirven L, et al. Diagnostic accuracy of MRI techniques for treatment response evaluation in patients with brain metastasis: a systematic review and meta-analysis. Radiother Oncol. 2022;177:121–33.

    Article  PubMed  Google Scholar 

  46. Eraky AM, Beck RT, Treffy RW, Aaronson DM, Hedayat H. Role of Advanced MR Imaging in diagnosis of neurological malignancies: current status and future perspective. J Integr Neurosci. 2023;22:73.

    Article  PubMed  Google Scholar 

  47. De Godoy LL, Chawla S, Brem S, Mohan S. Taming Glioblastoma in Real Time: integrating Multimodal Advanced Neuroimaging/AI tools towards creating a robust and Therapy Agnostic Model for Response Assessment in Neuro-Oncology. Clin Cancer Res. 2023;29:2588–92.

    Article  PubMed  Google Scholar 

  48. Siakallis L, Sudre CH, Mulholland P, Fersht N, Rees J, Topff L, et al. Longitudinal structural and perfusion MRI enhanced by machine learning outperforms standalone modalities and radiological expertise in high-grade glioma surveillance. Neuroradiology. 2021;63:2047–56.

    Article  PubMed  PubMed Central  Google Scholar 

  49. Müller SJ, Khadhraoui E, Voit D, Riedel CH, Frahm J, Ernst M. First clinical application of a novel T1 mapping of the whole brain. Neuroradiol J. 2022;35:684–91.

    Article  PubMed  PubMed Central  Google Scholar 

Download references

Acknowledgements

None.

Funding

Open Access funding enabled and organized by Projekt DEAL. We received no funding for the conduct of this study. We acknowledge support by the Open Access Publication fund of medical faculty of the Otto-von-Guericke-University Magdeburg regarding publication costs.

Author information

Authors and Affiliations

  1. Clinic for Neuroradiology, Otto-Von-Guericke-University Magdeburg, Leipziger Str. 44, D-39120, Magdeburg, Germany

    Eya Khadhraoui, Leon Schmidt, Stefan Klebingat, Roland Schwab, Harald Paukisch, Daniel Behme & Sebastian Johannes Müller

  2. Department of Neurological Surgery, Göttingen University Hospital, Robert-Koch-Str. 40, D-37075, Göttingen, Germany

    Silvia Hernández-Durán

  3. Clinic for Neuroradiology, Katharinen-Hospital Stuttgart, Kriegsbergstr. 60, D-70174, Stuttgart, Germany

    Georg Gihr

  4. Department of Neurosurgery, Otto-Von-Guericke-University Magdeburg, Leipziger Str. 44, D-39120, Magdeburg, Germany

    Klaus-Peter Stein

  5. Stimulate Research Campus Magdeburg, Otto-Hahn-Str. 2, D-39106, Magdeburg, Germany

    Daniel Behme

Authors
  1. Eya Khadhraoui
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Leon Schmidt
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Stefan Klebingat
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Roland Schwab
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Silvia Hernández-Durán
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Georg Gihr
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Harald Paukisch
    View author publications

    You can also search for this author in PubMed Google Scholar

  8. Klaus-Peter Stein
    View author publications

    You can also search for this author in PubMed Google Scholar

  9. Daniel Behme
    View author publications

    You can also search for this author in PubMed Google Scholar

  10. Sebastian Johannes Müller
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

All authors have reviewed and approved the submitted manuscript for publication. EK and SM were the project administrators and organized the data curation, measurements, design, literature review, and writing of the manuscript. SK programmed the Python software. LS contributed to data collection, measurements, and analysis. SH and GG contributed to the literature review and design of the study. HP, DB and RS reviewed the draft, and SM, KS and DB contributed to the formal analysis and literature review of the study, draft, review, and edited the manuscript. The authors agree to be accountable for all aspects of work ensuring integrity and accuracy.

Corresponding author

Correspondence to Sebastian Johannes Müller.

Ethics declarations

Ethics approval and consent to participate

The institutional review board of the University of Magdeburg (Leipziger Str. 44, D-39104 Magdeburg, Germany, No. 103/24) ethically approved this study, which adhered to the 2013 Helsinki Declaration. Because of the retrospective, observational nature of the study, the ethics committee waived the requirement for informed consent.

Consent for publication

Not applicable.

Competing interests

SM has received a speaker honorarium from BrainLab AG (Olof-Palme-Str. 9, D-81829 Munich, Germany) for lecture regarding Treatment Response Assessment Maps. The other authors declare that they have no competing interests.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Khadhraoui, E., Schmidt, L., Klebingat, S. et al. Comparison of a new MR rapid wash-out map with MR perfusion in brain tumors. BMC Cancer 24, 1139 (2024). https://doi.org/10.1186/s12885-024-12909-z

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s12885-024-12909-z

Keywords

BMC Cancer

ISSN: 1471-2407

Contact us