Multi-Dynamic-Multi-Echo-based MRI for the Pre-Surgical Determination of Sellar Tumor Consistency: a Quantitative Approach for Predicting Lesion Resectability

Purpose Pre-surgical information about tumor consistency could facilitate neurosurgical planning. This study used multi-dynamic-multi-echo (MDME)-based relaxometry for the quantitative determination of pituitary tumor consistency, with the aim of predicting lesion resectability. Methods Seventy-two patients with suspected pituitary adenomas, who underwent preoperative 3 T MRI between January 2020 and January 2022, were included in this prospective study. Lesion-specific T1-/T2-relaxation times (T1R/T2R) and proton density (PD) metrics were determined. During surgery, data about tumor resectability were collected. A Receiver Operating Characteristic (ROC) curve analysis was performed to investigate the diagnostic performance (sensitivity/specificity) for discriminating between easy- and hard-to-remove by aspiration (eRAsp and hRAsp) lesions. A Mann-Whitney-U-test was done for group comparison. Results A total of 65 participants (mean age, 54 years ± 15, 33 women) were enrolled in the quantitative analysis. Twenty-four lesions were classified as hRAsp, while 41 lesions were assessed as eRAsp. There were significant differences in T1R (hRAsp: 1221.0 ms ± 211.9; eRAsp: 1500.2 ms ± 496.4; p = 0.003) and T2R (hRAsp: 88.8 ms ± 14.5; eRAsp: 137.2 ms ± 166.6; p = 0.03) between both groups. The ROC analysis revealed an area under the curve of 0.72 (95% CI: 0.60–0.85) at p = 0.003 for T1R (cutoff value: 1248 ms; sensitivity/specificity: 78%/58%) and 0.66 (95% CI: 0.53–0.79) at p = 0.03 for T2R (cutoff value: 110 ms; sensitivity/specificity: 39%/96%). Conclusion MDME-based relaxometry enables a non-invasive, pre-surgical characterization of lesion consistency and, therefore, provides a modality with which to predict tumor resectability. Supplementary Information The online version of this article (10.1007/s00062-024-01407-1) contains supplementary material, which is available to authorized users.


Introduction
Pituitary adenomas account for approximately 15% of primary intracranial tumors [1].Although mostly benign, these lesions may show a tendency to affect anatomically adjacent structures in an invasive manner [1,2].Therefore, early neurosurgical interventions in certain cases are considered an effective treatment option that results in an auspicious clinical outcome [3].Since the endoscopic, transnasal, transsphenoidal route has become the most widely used approach for surgical removal of pituitary tumors, a pre-surgical understanding of an individual's anatomical complexity and lesion characteristics have become paramount [4].
While most pituitary adenomas can be easily extracted using this neurosurgical approach, 10-15% of these tumors are composed of fibrotic tissue, which are technically more demanding with regard to resection and require different neurosurgical techniques, as well as augmented resection equipment [5][6][7].Therefore, pre-surgical knowledge of tumor consistency is crucial and will facilitate neurosurgical planning, which is key to improve postoperative clinical outcomes [2,[8][9][10].However, modalities with which to assess pituitary adenoma consistency non-invasively prior to surgery are currently lacking [5,11].
Multi-dynamic-multi-echo (MDME)-based imaging generates MRI contrasts based on tissue-specific properties (i.e., longitudinal and transverse relaxation times), acquired within a single scan of less than six minutes, by retrospective modulation of repetition time and echo time (i.e., image synthesis) [12][13][14][15][16][17].Moreover, MDMEbased sequences enable the quantification of tissue-specific relaxation time properties and proton density metrics [13,16,17].While this novel modality has been investigated in various fields of clinical neuroradiology, studies that investigate lesion consistency using relaxometry are still scarce.Furthermore, only very few studies have focused on the applicability of contrast-enhanced MDME sequence acquisitions in a clinical setting [12,[18][19][20].
The aim of this prospective study was to investigate the feasibility of MDME-based MRI for the prediction of intraoperative resectability in a cohort of patients with suspected pituitary macroadenomas based on MRI evaluation.For this purpose, the lesion consistency was quantified by determining tissue-specific MR properties [T1-/T2-relaxation times (T1R/T2R) and proton density (PD) metrics] on non-enhanced, MDME-based imaging data.Lesion-specific MR properties of hard-to-remove tumors were compared to those considered easy-to-remove by aspiration during neurosurgery.In addition, relaxometry-based sensitivity/ specificity characteristics and their respective cutoff values regarding lesion texture were analyzed.Furthermore, the feasibility of contrast-enhanced, MDME-based T1weighted contrasts for the preoperative characterization of parasellar anatomy/pituitary adenomas was investigated.The results were compared to standard-of-reference T1-weighted sequences [Magnetization Prepared Rapid Acquisition Gradient Echo (MPRAGE) and Volumetric Interpolated Brain Examination (VIBE)].

Ethical Approval
The protocol of this prospective study was approved by the local ethics commission (Medical University of Vienna, Vienna, Austria) and performed in accordance with the Fig. 1 The flow diagram indicates how the study sample was derived.Patients in whom coronal multi-dynamic-multi-echo (MDME)-based imaging data were not available were excluded from this study (1/72).Furthermore, subjects who did not undergo neurosurgery were excluded from the quantitative analysis (6/72).Severe motion-related artifacts on post-contrast, MDME-based and Volumetric Interpolated Brain Examination (VIBE)-based imaging data were excluded from the qualitative analysis (1/72).In total, 65 participants ["easy-to-remove by aspiration" (eRAsp) = 41; "hard-to-remove by aspiration" (hRAsp) = 24] were included in the quantitative analysis, while 23 participants were enrolled in the qualitative assessment K * Data were compared using the Pearson χ 2 test for group homogeneity and the unpaired t-test a According to World Health Organization (WHO) 2017 [32] b Chondrosarcoma presenting as pituitary adenoma [33] c Intrasellar meningioma mimicking pituitary adenoma [34] d Intrasellar plasmacytoma presenting as pituitary macro-adenoma [35] eRAsp easy-to-remove by aspiration, hRAsp hard-to-remove by aspiration, LHC Langerhans cell histiocytosis, MRI Magnetic resonance imaging Helsinki Declaration of 1975.All patients provided written, informed consent prior to MRI scanning and agreed to the scientific use of the acquired imaging data (EC-Nr.: 1549/2019).

Study Cohort
Between January 2020 and January 2022, pre-surgical MRI was performed in 72 patients with a tumor in the pituitary region at the Department of Neuroradiology of a tertiary care hospital.All lesions have been morphologically delineated as pituitary macroadenomas based on MRI evaluation.Pituitary adenoma was histologically confirmed in 61 of the 72 patients (Table 1).Histological data were not available in 2 patients.Microadenoma patients were excluded from this investigation.Study sample characteristics are shown in Fig. 1 and Table 1.

MR Data Acquisition
MRI was performed using the institutional routine imaging protocol for pre-neurosurgical planning (Table 2).Pre-contrast, MDME-based sequences (coronal plane) were added to the protocol.In 24 patients, post-contrast, MDME-based T1-weighted data were also available.The imaging data were acquired on a Siemens MAGNETOM Vida (3 T) MR system.Via two repeated acquisition phases [phase a, a slice-selective saturation pulse (flip angle: 120°) was applied to saturate one slice; phase b, a series of sliceselective refocusing pulses (flip angle: 180°) and a sliceselective excitation pulse (flip angle: 90°) were applied to generate spin echoes for another slice], the MDME sequence derived information about the physical MR properties of the tissue (T1-/T2-relaxation parameters and spin density) [13,14,21].Data acquisition was performed at two different echo times (24 ms and 107 ms) following the 90°pulses, which were applied at four different saturation recovery times throughout the MDME scan cycle (i.e.,

Determination of Lesion-Specific MR Properties
Prior to the analysis, a critical visual review of the acquired MR data was performed by one neuroradiologist (W.M.) with 15 years of experience in neuroimaging.Highly artifact-degraded acquisitions were excluded.The lesion-specific properties [T1R (ms); T2R (ms); PD (%)] were measured by manually placing regions of interest (ROIs) at three different slices of the tumor on pre-contrast, "SyMRI"-generated parametric maps (Fig. 2).Units for T1R (spin-latticerelaxation), T2R (spin-spin-relaxation), and PD (number of nuclei in the area being imaged) were adopted as provided by the "SyMRI"-based default software settings.ROI positioning was performed by two independent raters (M.S.Y. and L.L., each with one year of experience in pituitary MRI), who were blinded to intraoperative resectability assessments.In four cases, only two ROIs were drawn due to the limited availability of slices of the lesion.The average values of the quantitative properties were automatically computed, based on each single voxel value within the drawn ROI.Mean values, based on the three ROI measurements in each lesion, were calculated and used for further analysis (Supplementary Table 1).

Assessment of Intraoperative Resectability of Sellar Lesions
The assessment of the intraoperative tumor extractability, via an endoscopic, transnasal, transsphenoidal approach, was assessed by the attending neurosurgeons (A.M. and S.W.) during surgery.In contrast to soft-constituted adenomas, tumors composed of fibrotic components are considered difficult to resect [5].Therefore, the tumors were divided into two groups: "hard-to-remove by aspiration" (hRAsp) and "easy-to-remove by aspiration" (eRAsp).eRAsp lesions were freely removable by aspiration and only minimal use of curettage was required.Tumors that were difficult to aspirate, and which made curettage and successive mechanical debulking necessary, were assessed as hRAsp [5,22].

Qualitative Assessment of the Pituitary Region and Parasellar Anatomy
Qualitative assessments were performed on contrast-enhanced, MDME-based, VIBE-based, and MPRAGE-based T1-weighted data (coronal plane) (Fig. 3) by two independent senior neuroradiologists [J.F.(15 years of experience in neuroimaging) and W.M.].To evaluate the pituitary region and parasellar anatomy, the following criteria were used [23]: the Knosp grade (for left/right cavernous sinus invasion) [23,24]; parasellar invasiveness; normal gland position; detectability of the optic chiasm and left/right oculomotor nerve.The scoring criteria are explained in Supplementary Table 2 and Supplementary Fig. 1.Supplementary Fig. 2 shows the differences between pre-and post-contrast MDME-based acquisitions.

Statistical Analyses
The statistical analyses were proposed and performed by one biomedical statistician with 30 years of experience K A p-value of p ≤ 0.05 was considered statistically significant.An intra-class correlation coefficient (ICC) analysis was used to assess the overall agreement between the mean values determined by both raters.ICC values ≥ 0.75 were considered strong agreements [25].In case of strong correlations, the data based on the determinations performed by rater 1 were used for further analysis.A Mann-Whitney-Utest was performed to detect differences in T1R, T2R, and PD metrics between both groups (hRAsp vs. eRAsp).Receiver Operating Characteristic (ROC) curve analyses were performed to evaluate the predictive power of quantitative MRI metrics regarding the pre-surgical discrimination between hRAsp and eRAsp lesions.The Youden indices were calculated to determine optimal cutoff values for each quantitative metric.
In 24 of the 72 patients, contrast-enhanced, MDMEbased T1-weighted sequence acquisitions were also available.One of the 24 patients with contrast-enhanced acquisitions had to be excluded from the qualitative analysis, due to severe motion-related artifacts that were visible only on the contrast-enhanced, MDME-based and VIBE-based T1weighted MR imaging data.Thus, for the qualitative subanalysis of the pituitary region/parasellar anatomy, a total of 23 participants (mean age at data acquisition, 52 years ± 18) (female/male: 12/11) were included in this study (Table 1).

Discussion
In this study, the feasibility of multi-dynamic-multi-echo (MDME)-based MRI for the quantitative and qualitative pre-neurosurgical characterization of lesions, which have been identified as pituitary macroadenomas on pre-surgical MR imaging, was investigated in a clinical setting.Relaxometry-based mapping enabled the non-invasive assessment of lesion consistency and, therefore, provided an easy-to-apply modality with which to predict intraoperative tumor resectability [T1-relaxation time (T1R): sensitivity/specificity 78%/58%; cutoff value of 1248 ms (AUC = 0.72) and T2-relaxation time (T2R): sensitivity/specificity 39%/96%; cutoff value of 110 ms (AUC = 0.66)].Moreover, the results presented in this investigation suggest that con-  Most pituitary adenomas are characterized by a soft lesion consistency, and, therefore, are easily removed using aspiration devices via minimally invasive, transnasal, transsphenoidal approaches.However, pituitary lesions that contain fibrous components may be difficult to extract using the aforementioned approach and require different neurosurgical techniques for removal [5,7].Pituitary adenomas with a fibrous component account for approximately 10-15% of sellar lesions [5][6][7].These tumors are associated with lower total resection rates and higher risks of recurrence after surgery, which is accompanied by an unsatisfactory clinical outcome [2,[8][9][10].Thus, a priori information about lesion texture may facilitate preoperative planning with the chance to improve post-surgical outcomes.
MDME-based imaging is considered a relatively novel MR approach, which excels because of the short examination time and the possibility to retrieve data for both quantitative and qualitative evaluations.Although there are several studies that have focused on the clinical applicability of this recent technology, there is a lack of information on the practicality of MDME-derived data for neurosurgical needs [12,[18][19][20].Quantitative MR metrics are linked to tissue-specific properties.Primarily, the H2O/protein fraction of the tissue determines its relaxometric features, with increased relaxation times/PD metrics associated with higher water content and vice versa [27].These considerations are in line with the presented data, since lesions classified as eRAsp revealed higher T1R, T2R, and PD metrics, while diametrically opposed results were observed for hRAsp tumors.Furthermore, our observations are in keeping with a previous study by Yamada et al., who demonstrated similar K findings in a cohort of patients with meningiomas, based on a different MR mapping approach [28].Thus, quantitative imaging modalities appear to provide robust biomarkers for the determination of lesion consistency, despite potential differences in mapping technology [29].
Nonetheless, currently, contrast-enhanced T1-weighted imaging, for the qualitative assessment, represents the mainstay prior to neurosurgery.The feasibility of post-contrast MDME-based T1-weighted contrasts was investigated using a scoring system that evaluated crucial anatomical aspects of pituitary surgery.MDME-derived data demonstrated non-inferiority to MPRAGE-based data.While almost excellent concordances were observed for nearly all evaluated aspects, inter-rater (MDME/MPRAGE) and inter-sequence (MDME vs. MPRAGE) agreement was lowest for the detectability of the left oculomotor nerve, most likely due to higher Knosp grades assigned for the left side.Interestingly, apart from lower inter-rater concordances for oculomotor nerve detection based on VIBE sequences, there was relatively low agreement the detectability of the optic chiasm compared to MDME-and MPRAGEbased data, possibly explained by the fact that higher resolution MR acquisitions are more prone to motion-related artifacts [30].Nonetheless, high-resolution MRI remains indispensable for pre-surgical planning [31].
MDME-based imaging provides the opportunity to supply multi-parametric characterizations of the tissue to be resected and enables investigators to retrieve reliable, postcontrast T1-weighted data for the anatomical assessment prior to surgery.Moreover, the presented approach provides the opportunity to reconstruct various MR contrasts based on a single scan, which may be of interest in a neurosurgical setting.However, this was beyond the scope of this work.Nonetheless, the investigated modality bears promising potential to aid in neurosurgical decision-making and may improve preoperative planning.

Limitations
Several limitations require consideration.We included all pituitary tumors initially classified as macroadenomas based on MR imaging.Therefore, based on histology, various subtypes of macroadenomas were included.Furthermore, there were also other solitary tumor entities, mimicking macroadenomas, that were included in this study.The sample size for both quantitative and qualitative analyses was relatively small, which mandates the need for further studies to confirm our findings.Furthermore, there was a considerable delay between intravenous contrast agent administration and MDME-based sequence acquisitions for the qualitative sub-analysis.However, the acquired data proved sufficient to reliably study the applicability of MDME-based imaging data in a clinical setting.This study did not provide information on the feasibility of MDMEbased imaging for the assessment of pituitary microadenomas since these require different imaging acquisition strategies [32].Nonetheless, the relaxometric evaluation of pituitary microadenomas is of great interest and requires further consideration in the future.

Conclusion
In summary, multi-dynamic-multi-echo (MDME)-based mapping represents a reliable method with which to predict the intraoperative resectability of sellar tumors by providing non-invasive biomarkers for lesion consistency.Moreover, synthetically generated, contrast-enhanced T1weighted data approached a performance similar to that of the current standard-of-reference with regard to presurgical assessments of the pituitary region.Therefore, the presented imaging approach provides promising potential to aid in neurosurgical decision-making and to facilitate preoperative planning, which is key to improved neurosurgical performance and post-procedural outcomes.This investigation paves the way for multi-parametric, MDMEbased mapping in clinical neurosurgery.

Fig. 3
Fig. 3 multi-dynamic-multi-echo-based (a); Volumetric Interpolated Brain Examination (VIBE)-based (b); and Magnetization Prepared Rapid Acquisition Gradient Echo (MPRAGE)-based (c) T1-weighted imaging data (coronal plane) depict parasellar anatomy comparably in a female patient, 22 years of age at the time of MRI (non-invasive, lactotroph adenoma).The left and right oculomotor nerve (arrows) is detectable on each sequence acquisition.Notice the differences in carotid artery enhancement (asterisk), while the pituitary adenoma enhances in a relatively stable manner, despite protocol-related delays (Table2) Characteristic curve analysis to investigate diagnostic performance and Mann-Whitney-U-test AUC Area under the curve, eRAsp easy-to-remove by aspiration, hRAsp hard-to-remove by aspiration, ICC Intra-class correlation coefficient, MRI Magnetic resonance imaging, PD proton density, T1R T1-relaxation time, T2R T2-relaxation time 0.85 (95% CI: 0.75-0.91,p < 0.001)] as determined by both raters.
09-0.88) b a Inter-sequence Agreement (κ) for Observer 1 b Inter-sequence Agreement (κ) for Observer 2 Numbers in parentheses are 95% confidence intervals CN Cranial nerve, MDME Multi-dynamic-multi-echo, MPRAGE Magnetization Prepared Rapid Acquisition Gradient Echo, VIBE Volumetric Interpolated Brain Examination trast-enhanced, MDME-based T1-weighted contrasts enable robust pre-surgical evaluations similar to those provided by Magnetization Prepared Rapid Acquisition Gradient Echo (MPRAGE), standard-of-reference MR sequences.

Table 1
Characteristics of Included Participants

Table 2
Magnetic Resonance Imaging Protocol MPRAGE Magnetization Prepared Rapid Acquisition Gradient Echo, TE Echo time, TR Repetition time, TOF Time-of-flight angiography, VIBE Volumetric Interpolated Brain Examination eight spin echo acquisitions per scan cycle).Quantitative maps for the determination of lesion-specific properties were generated using the MDME post-processing soft- AT Acquisition time, CISS Constructive interference in steady state, FOV Field-of-view, MDME Multi-dynamic-multi-echo, ware "SyMRI" (Synthetic MR AB, Linköping, Sweden; Version 11.2.9).Generation of parametric maps was performed within approximately ten seconds (per imaging data set), after transfer of MDME-based acquisitions from the MR scanner to a separate workstation for quantitative analysis.Contrast-enhanced, MDME-based T1-weighted data were generated automatically and retrieved on a Picture Archiving and Communication System (PACS) workstation for qualitative pre-surgical assessment.

Table 3
Descriptive Statistics and Receiver Operating Characteristic Curve Analysis