Introduction
Meningiomas are the most common primary intracranial neoplasms in adults, accounting for more than 30% of all intracranial neoplasms.1,2 According to the World Health Organization (WHO) grading system, meningiomas are classified into three grades based on specific histological criteria: WHO grade 1 (benign), WHO grade 2 (atypical), and WHO grade 3 (malignant).3,4 Grade 2 meningiomas are defined by increased mitotic activity and/or distinctive histological features. Clear cell and chordoid subtypes are also classified as grade 2 meningiomas. In the 2007 and 2016 WHO classifications, brain invasion was included as an independent criterion for grade 2 (atypical) meningiomas. This revision in classification has increased the proportion of grade 2 meningiomas from 7% to 15–20%.5–7
Surgical resection, aiming for maximal safe removal of the tumor, remains the primary treatment for grade 2 meningiomas. However, unlike benign meningiomas, grade 2 meningiomas exhibit a significant tendency for recurrence, even after radical surgical resection with or without adjuvant radiotherapy (RT). Stereotactic radiosurgery (SRS) has increasingly been used in selected patients for residual or recurrent grade 2 meningiomas as an alternative or adjunct to external beam radiotherapy (EBRT), offering comparable local control rates.8–11 Despite these advancements, the specific role of radiosurgery in the treatment of grade 2 meningiomas remains unclear. This uncertainty stems from the heterogeneity of patient populations, complex treatment histories, small sample sizes in studies, and the evolving WHO diagnostic criteria for grade 2 meningiomas. A recent multicenter retrospective study examined a large cohort of patients treated with gamma knife radiosurgery (GKRS) for WHO grade 2 meningiomas, providing data on survival rates, progression-free survival (PFS), and the incidence of adverse radiation events. However, this study did not describe the patterns of tumor recurrence, specifically whether recurrence occurred within or outside the irradiation field.12 In this study, we evaluated the long-term outcomes of a large series of patients with grade 2 meningiomas treated with GKRS. We analyzed factors affecting tumor control, patterns of recurrence, and survival to further elucidate the role of GKRS in managing these aggressive tumors.
Materials and methods
We retrospectively reviewed records for all patients who underwent GKRS for grade 2 meningioma between January 2007 and December 2016 at Beijing Tiantan Hospital. Patients with neurofibromatosis type 2 were excluded from the study. This study was carried out in accordance with the recommendations of the Declaration of Helsinki (as revised in 2013). This study was approved by Beijing Tiantan Hospital institutional review board (Approval No. KY2020-135-01). The individual consent for this retrospective analysis was waived.
Baseline demographic variables were documented for each patient, including age, gender, tumor location, prior radiation therapy, prior surgical resections, presence of single versus multiple lesions, prior recurrence, and treatment characteristics. A total of 75 patients were identified for this study, but five patients without subsequent radiological follow-up were excluded from the analysis.
Patient demographics
The patient characteristics are summarized in Table 1. Seventy patients were included in this cohort study, comprising 39 females (55.7%) and 31 males (44.3%). The median age was 46 years (range: 16–77 years). These 70 patients had 79 separate foci of grade 2 meningioma treated with GKRS; 63 patients had a single tumor, and seven patients had multiple tumors. Prior RT was administered in 11 patients, including EBRT in nine patients and GKRS in two patients. The median number of prior recurrences was one (range: 0–6). Twenty-four patients underwent adjuvant GKRS for residual tumors at a median of four months (range: 2–6 months) after the initial surgery. Forty-six patients underwent salvage GKRS at a median of 37 months (range: 6–120 months) following initial surgery. Of these 46 patients, 32 experienced a single recurrence at a median of 18 months (range: 6–84 months), while 14 experienced multiple recurrences at a median of 60 months (range: 14–120 months).
Table 1Patient and treatment characteristics
| Variable | Value |
|---|
| Sex, M/F | 31/39 |
| Median age, y [range] | 51 [16–77] |
| Prior surgery resections median [range] | 1 [1–4] |
| 1 | 54 |
| 2 | 12 |
| 3 | 2 |
| 4 | 2 |
| Prior radiation therapy | 11 |
| External beam radiation therapy (54–56 Gy) | 9 |
| Gamma knife (12–15 Gy) | 2 |
| Tumor location | |
| Convexity | 11 |
| Falx/parasagittal | 17 |
| Skull base | 42 |
| Single vs. multiple lesions | |
| Single | 63 |
| Multiple | 7 |
| Prior recurrence | |
| Group1: adjuvant therapy for residual tumor after initial surgery | 24 |
| Group2: salvage therapy for tumor at first recurrence | 32 |
| Group3: salvage therapy for tumor at multiple recurrences | 14 |
| Treatment characteristic | |
| Median maximum dose, Gy [range] | 28 (18–33) |
| Median peripheral dose, Gy [range] | 13 (9–17) |
| Median isodose, [range] | 50 (40–60) |
| Median tumor volume, CM3 [range] | 3.9 (0.15–29.9) |
Radiosurgical parameters
A Leksell stereotactic frame was affixed to each patient’s skull under local anesthesia. High-resolution, contrast-enhanced magnetic resonance imaging (MRI) with a slice thickness of 2 mm was performed for treatment planning. Patients underwent single-fraction SRS using the Leksell Gamma Knife Model C (Elekta AB) from January 2007 to October 2011, after which the system transitioned to the Leksell Gamma Knife Perfexion (Elekta AB). Treatment planning was conducted using the GammaPlan system (Elekta AB). The prescription dose and isodose lines were selected based on proximity to critical structures, prior radiation therapy, and tumor size. The median prescription dose was 13 Gy (range: 9–17 Gy) delivered to a median isodose line of 50.0% (range: 40–60%). The median maximum dose was 28 Gy (range: 18–33.3 Gy). The median tumor volume was 3.9 cm3 (range: 0.15–29.9 cm3).
Follow-up
Patients were advised to undergo MRI and clinical evaluations every six months during the first year after GKRS, followed by intervals of six to twelve months thereafter. Tumor recurrence was defined as an increase in tumor size or the appearance of a new tumor on follow-up MRI after GKRS. Recurrence was classified as local recurrence (tumor enlargement after treatment), marginal recurrence (new tumor outside the target area but within or immediately adjacent to the resection cavity), or distant recurrence (new tumor in distant locations). To minimize variability between scanners and images, tumor shrinkage or growth was defined as a 25% decrease or increase in volume, respectively.
Statistical analysis
The Kaplan–Meier method was used to estimate overall survival (OS), PFS, local control (LC), marginal control (MC), and distant control (DC). OS, PFS, LC, MC, and DC times were calculated from the date of the first GKRS at our hospital. Only the first instance of each recurrence pattern was considered. PFS was calculated per patient, while LC was calculated per lesion. Univariate analysis was performed using the log-rank test for the following factors: sex, age (≤51 vs. >51 years), prior recurrence (Group 1: adjuvant therapy for residual tumors after initial surgery; Group 2: salvage therapy for tumors at first recurrence; Group 3: salvage therapy for tumors at multiple recurrences), tumor volume (≤13 vs. >13 cm3), margin dose (≤13.0 vs. >13 Gy), maximum dose (≤28 vs. >28 Gy), prior RT (yes vs. no), and tumor location (parasagittal/convexity vs. skull base). Factors with a p-value ≤ 0.05 in univariate analysis were entered into multivariate Cox regression analysis. A p-value ≤ 0.05 was considered statistically significant. All statistical analyses were performed using SPSS version 22.0 (IBM Corporation, Armonk, NY, USA).
Results
Tumor control and survival
In the current series, the median duration of follow-up was 48 months (range: 8–132 months). Of these 79 foci, follow-up MRI showed that 33 foci (42%) remained stable, 22 foci (28%) decreased, and 24 foci (30%) increased. The total tumor control rate was 70%.
At the time of the last follow-up, tumor recurrence was demonstrated in 39 of 70 (56%) patients. Fifteen patients (21%) died during a median period of 34.5 months (range: 8–72 months) after GKRS, of whom 14 died due to tumor progression and 1 due to an unknown cause.
The occurrence of tumor recurrence was documented as follows: local recurrence in ten patients, marginal recurrence in fifteen patients, distant recurrence in one patient, local + marginal recurrence in eight patients, marginal + distant recurrence in one patient, local + distant recurrence in one patient, and local + marginal + distant recurrence in three patients. Survival curves were calculated to illustrate the time to death and recurrence. The OS rates at one, three, and five years after GKRS were 99%, 88%, and 77%, respectively. The one-, three-, and five-year LC rates were 92%, 73%, and 65%, respectively. The one-, three-, and five-year MC rates were 94%, 63%, and 56%, respectively. The one-, three-, and five-year DC rates were 96%, 92%, and 90%, respectively. The one-, three-, and five-year PFS rates were 87%, 51%, and 44%, respectively.
Risk factors for tumor progression
Univariate analysis using the log-rank test was performed to assess potential risk factors for OS, LC, MC, DC, and PFS (Tables 2, 3; Figs. 1, 2). All factors with p ≤ 0.05 in the univariate analysis were entered into multivariate Cox regression analysis. Univariate analysis revealed that factors associated with worse OS included age > 51 years (p = 0.005), multiple prior recurrences (p < 0.001), parasagittal/convexity lesions (p = 0.025), prior radiation therapy (p = 0.008), and multiple lesions (p < 0.001). Multivariate analysis showed that multiple prior recurrences (p = 0.004) and age > 51 years (hazard ratio [HR] = 4.0, p = 0.04) were independent factors influencing survival.
Table 2Univariate and multivariate factors associated with MC, DC, and LC
| Factors | Marginal control
| Distant control
| Local control
|
|---|
| Univariate p-value | Multivariate HR, 95% CI,p-value | Univariate p-value | Multivariate HR, 95% CI (p-value) | Univariate p-value | Multivariate HR, 95% CI (p-value) |
|---|
| Age (years) | 0.274 | – | 0.009 | – | 0.396 | – |
| Gender | 0.102 | – | 0.551 | – | 0.172 | – |
| Multiple or single | 0.003 | 3.9,1.4–10.8 (0.008) | 0.004 | – | 0.002 | 4.1,1.6–10.9 (0.004) |
| Tumor volume | 0.889 | – | 0.283 | – | 0.889 | – |
| Margin dose | 0.251 | – | – | – | 0.012 | 3.2,1.2–8.3 (0.016) |
| Maximum dose | 0.194 | – | – | – | 0.147 | – |
| Prior recurrence | 0.004 | p = 0.01 | p < 0.001 | – | 0.131 | – |
| Group(1) | | 0.2,0.1–0.6 (0.004) | | – | | |
| Group(2) | | 0.3,0.1–0.8 (0.018) | | – | | |
| Group(3) | | reference | | – | | |
| Prior RT | 0.124 | – | p < 0.001 | – | 0.762 | – |
| Tumor location | 0.402 | – | 0.338 | – | 0.073 | – |
Table 3Univariate and multivariate factors associated with PFS and OS
| Factors | Progression free survival
| Overall survival
|
|---|
Univariate
p-value | Multivariate HR, 95% CI (p-value) | Univariate p-value | Multivariate HR, 95% CI (p-value) |
|---|
| Age (years) | 0.381 | – | 0.005 | 4.0,1.1–15.0 (0.04) |
| Gender | 0.159 | – | 0.348 | |
| Multiple or single | 0.008 | 2.7,1.0–7.0 (0.04) | p < 0. 001 | 4.6,1.0–22.1 (0.054) |
| Tumor volume | 0.485 | – | 0.597 | |
| Margin dose | 0.039 | 2.0,1.0–4.3 (0.066) | 0.087 | |
| Maximum dose | 0.522 | – | 0.891 | |
| Prior recurrence | 0.001 | p = 0.015 | p < 0.001 | p = 0.04 |
| group(1) | | 0.2,0.1–0.6 (0.004) | | 0.11,0.02–0.59 (0.10) |
| group(2) | | 0.3,0.1–0.8 (0.017) | | 0.09,0.02–0.43 (0.002) |
| group(3) | | – | | |
| Prior RT | 0.02 | 0.7,0.2–2.4 (0.617) | 0.008 | 1.0,0.2–4.3 (1.00) |
| Tumor location | 0.08 | – | 0.025 | 2.3,0.8–7.1 (0.14) |
In univariate analysis, multiple lesions (p = 0.002) and margin doses ≤ 13 Gy (p = 0.012) were predictive of worse local control. There was also a trend toward worse local control with parasagittal/convexity lesions (p = 0.073). The mean peripheral doses were similar for sagittal sinus/convexity lesions (13.4 Gy) and skull-base lesions (13.3 Gy). A paired t-test did not reveal any statistically significant differences at the p < 0.05 level. In multivariate Cox regression analysis, multiple lesions (HR = 4.1, p = 0.004) and margin doses ≤ 13 Gy (HR = 3.2, p = 0.016) were independent negative predictors of local control.
In univariate analysis, multiple lesions (p = 0.003) and multiple prior recurrences (p = 0.004) were predictive of worse marginal control. In multivariate Cox regression analysis, multiple lesions (HR = 3.9, p = 0.008) and multiple prior recurrences (p = 0.01) were independent negative predictors of marginal control.
In univariate analysis, multiple lesions (p = 0.004), multiple prior recurrences (p < 0.001), age > 51 years (p = 0.009), and prior radiation therapy (p < 0.001) were predictive of worse distant control. Due to the limited number of patients with distant recurrence, multivariate analysis was not performed.
In univariate analysis, multiple lesions (p = 0.008), multiple prior recurrences (p = 0.001), margin doses ≤ 13 Gy (p = 0.039), and prior radiation therapy (p = 0.02) were predictive of worse PFS. In multivariate Cox regression analysis, multiple lesions (HR = 2.7, p = 0.04) and multiple prior recurrences (p = 0.015) were independent predictors of PFS.
There was no significant difference in OS (p = 0.758), local control (p = 0.618), marginal control (p = 0.389), distant control (p = 0.285), and PFS (p = 0.326) between Group 1 (adjuvant therapy for residual tumor after initial surgery) and Group 2 (salvage therapy for tumor at first recurrence) in the subgroup analysis.
Radiation-related complications and additional treatment
Three patients (4.3%) developed severe radiation-related complications after GKRS. Two patients experienced symptomatic radiation necrosis (worsening hemiparesis, n = 1; increased seizures, n = 1), which was treated with corticosteroids. Both patients had undergone multiple resections and EBRT before GKRS. The third patient had a 3.9 cm left temporal tumor treated with a marginal dose of 11 Gy. Subsequently, the patient developed increased seizures associated with edema and underwent gross total resection. This patient had no previous radiation therapy.
Of the 39 patients with tumor progression, 30 accepted further treatment after GKRS (Fig. 3). The other nine patients elected not to pursue additional treatments. During the follow-up period, 18 patients underwent 19 additional surgical resections for tumor progression (eight local recurrences, three marginal recurrences, four both local and marginal recurrences, one distant recurrence, and two with all three recurrence patterns). Twenty-three patients underwent 33 repeat SRS procedures for tumor progression (10 local recurrences, 19 marginal recurrences, two both local and marginal recurrences, and two distant recurrences). Additionally, two patients underwent adjuvant EBRT after reoperation. Of the 18 patients undergoing reoperation, five patients’ tumors (28%) showed progression from grade 2 to grade 3. The median time from GKRS to grade 3 meningioma was 36 months (range: 6–100 months).
Discussion
Grade 2 meningiomas are aggressive neoplasms with high recurrence rates. They have significantly worse outcomes after either recurrence or subtotal resection, approximating those of grade 3 meningiomas.11 SRS has also been increasingly used as an alternative therapeutic option to EBRT for patients with residual or recurrent grade 2 meningiomas.8–10 In this series, we demonstrate that GKRS can achieve an acceptable five-year local control rate (65%) and PFS rate (44%), and most patients tolerated GKRS well, except for three (4%) patients who had previous EBRT or large tumors. Furthermore, none of the 23 patients who received repeated GKRS after initial GKRS developed symptomatic adverse radiation effects.
In previous literature, the PFS rates at five years for grade 2 meningiomas treated with radiosurgery ranged from 16% to 83%, and the outcomes varied widely (Table 4).12–23 Grade 2 meningiomas represent a broad spectrum of tumors, and their biological behavior is highly heterogeneous. Due to this heterogeneity, small sample sizes can lead to large variances in results. Most previous studies consisted of relatively small sample sizes, which may partly account for this variance. Recently, a large multicenter study by Kowalchuk et al.12 analyzed 233 patients with grade 2 meningiomas treated with radiosurgery and reported PFS rates of 53.9% and 33.1% at three and five years, respectively. Symptomatic radiation effects occurred in 6% of patients. Our study has a relatively large sample size comparable to that of a previous single-center study, and our results are consistent with those of the multicenter study.
Table 4Literature summary of the main clinical studies on grade 2 meningioma treated with stereotactic radiosurgery
| Author (year) | Number of patients | Dose (Gy) | Volume (cm3) | Follow up (month) | Control rate (grade 2) |
|---|
| Harris et al.,200312 | II: 18 | mean 14.9 | mean 14.6 | 28 | five-year PFS:83% |
| Kano et al.,200713 | II: 10 III: 2 | mean 18 | median 2.87 | 43 | five-year PFS:48.3% |
| Attia et al.,201214 | II: 24 | median 14 | median 7.9 | 43 | five-year PFS:25% |
| Pollock et al.,201215 | II: 37 III: 13 | median 15 | median 14.6 | 38 | five-year PFS:40% |
| Hanakita et al.,201316 | II: 22 | median 18 | median 6 | 23.5 | five-year PFS:16% |
| Aboukais et al.,201517 | II: 27 | mean 15.2 | mean 5.4 | 56.4 | three-year LC:40%, three-year RC:33% |
| Wang et al.,201618 | II: 37 III: 9 | median 12.5 | median 11.7 | 32.6 | three-year PFS:30.6%, five-year PFS:20.4% |
| Zhang et al.,201619 | II: 44 | Median 20 | median 3.33 | 28 | three-year RC:58%, four- to five-year RC:58% |
| Reffat et al.,201720 | II: 75 | mean 16 | mean 3.5 | 41 | five-year LC:55.7% |
| Kowalchuk et al.,202121 | II: 233 | Median 15 | median 4.74 | 37.6 | three-year PFS:53.9%, five-year PFS:33.1% |
| Helis et al., 202022 | II, III: 48 | Median 15 | median 2.49 | 68.6 | five-year DFS:47.2% |
| Our study | II: 70 | median 13 | median 3.9 | 48 | three-year PFS:51%, five-year PFS:44% |
The prospective phase II trial (RTOG 0539) investigated the outcomes of intensity modulated radiotherapy (IMRT) with a dose of 60 Gy for high-risk meningiomas (newly diagnosed or recurrent grade 3 meningiomas, recurrent grade 2, or newly diagnosed grade 2 meningiomas after subtotal resection). The authors reported a three-year PFS of 58.8%, which was similar to our series with a three-year PFS rate of 51.4%.11 In our series, marginal recurrence was defined as new tumor formation outside the prescribed line but within the resection cavity. Our results showed high marginal recurrence comparable with IMRT (27/39 vs. 1/14). GKRS precisely delivers a high dose of radiation to the target and has a limited effect on outfield tissues. Considering the site of marginal recurrence, only marginal recurrence close to the prescribed line might be controlled by higher doses. Our analysis showed that margin dose was not associated with marginal recurrence. Valery et al.23 also analyzed marginal recurrence in grade 2 meningiomas and observed only a trend toward better marginal control with higher doses. To decrease the possibility of marginal recurrence, enlargement of the target volume should be considered, especially for small lesions or surgical beds. Another option might be EBRT with an SRS boost. Helis et al.22 reported six patients with grade 2 meningiomas who underwent SRS as a boost to EBRT. Of these six patients, none experienced marginal failure, and only one experienced local failure at the time of the last follow-up. However, two patients experienced severe complications after SRS. In our series, two of nine patients receiving prior EBRT had symptomatic radiation necrosis after GKRS. Given the heterogeneity of grade 2 meningiomas and radiation-related complications, we recommend SRS boosts in patients with risk factors associated with marginal recurrence. To reduce high-grade toxicity, Helis et al.22 also proposed that the SRS boost should target the areas immediately adjacent to the tumor cavity without expansion. This strategy may be a promising treatment option, but further investigation is needed.
In previous literature, factors associated with tumor progression and overall survival included age, tumor location, tumor size, mitotic index, prescription dose, timing of treatment, prior RT, conformality index, multiple lesions, and multiple prior recurrences.12–24 Among these factors, some reflect the aggressive nature of grade 2 meningiomas, while others are therapy-related parameters. Our results show that multiple lesions and multiple prior recurrences were significantly associated with worse PFS, marginal recurrence, and distant recurrence. Grade 2 meningiomas with these factors may represent inherently more aggressive lesions. Chen et al.24 reported 65 patients with atypical meningiomas who received salvage therapy, including surgery, RT alone, and surgery with adjuvant RT. They found that multifocal recurrences and more previous recurrences were associated with further progression. Helis et al.22 identified 48 patients who had 183 high-grade meningiomas treated with SRS and found that distant recurrences were significantly more common in patients with multiple lesions treated in a single SRS session. Several studies have demonstrated a positive correlation between increased dosage and enhanced tumor control. The recommended minimum doses are 12–18 Gy.15,17,19,23 Meng et al.25 recently reported 88 patients who underwent GKRS for grade 2 meningiomas in a single session and found that for tumors with Ki-67 > 10%, a margin dose of ≥14 Gy showed significantly better tumor control, but this was not the case for tumors with Ki-67 ≤ 10%. Our multivariate analysis indicated that a marginal dose of ≤13 Gy independently predicts local recurrence. The high-dose group demonstrated higher rates of local control (≥14 Gy: 86% at three years vs. ≤13 Gy: 64% at three years). We suggest that higher doses should be applied to achieve better local control. However, high-dose therapy may present challenges in certain situations, such as proximity to critical structures, prior radiotherapy, and large tumor sizes. Hypofractionated stereotactic radiosurgery has been utilized for large tumors or those close to critical structures due to its potential advantages. Marchetti et al.26 analyzed 24 patients with grade 2 meningiomas treated with multisession radiosurgery and reported a three-year PFS of 47%, a local control rate of 86%, and an adverse radiation effect rate of 4%. However, additional investigations are needed to confirm the long-term outcomes of hypofractionated stereotactic radiosurgery for grade 2 meningiomas.
The optimal timing of GKRS is not well established. Our results show that patients receiving salvage therapy for tumors at multiple recurrences had significantly worse outcomes than those receiving adjuvant therapy for residual tumors after initial surgery or salvage therapy for tumors at first recurrence. Because multiply recurrent meningiomas are more aggressive, a comparison between patients receiving SRS after initial resection and those receiving SRS at first recurrence is appropriate to evaluate the optimal timing of SRS. In our subgroup analysis, there was no significant difference in overall survival or tumor control between adjuvant therapy for residual tumors after initial surgery and salvage therapy for tumors at first recurrence. However, large tumor volumes, associated with worse PFS, have been reported in previous series.15,16,21 The enlargement of the tumor may lead to lower prescription doses and increased morbidity. Therefore, we suggest that residual tumors should be treated aggressively within a short interval.
This study was retrospective and had its limitations. Rigorous analysis of radiation-related complications was particularly difficult because many patients experienced multiple relapses and repeated treatments. However, based on clinical and radiographic findings, the assessment of severe radiation-related complications was reliable. Future directions to improve the treatment of grade 2 meningiomas may stem from recent advancements in next-generation sequencing, which may better distinguish the heterogeneity of grade 2 meningiomas. Molecular classification of meningiomas can more accurately reflect the biological behavior of the tumor and improve the selection of treatment parameters.
Conclusions
Our results support the use of GKRS as a reasonable treatment option in the management of grade 2 meningiomas. A higher margin dose should be considered to achieve better local control. Outfield progression (marginal and/or distant recurrence) was common, particularly in patients with multiple prior recurrences and/or multiple lesions. More aggressive treatment strategies should be explored for patients with these risk factors.
Declarations
Ethical statement
This study was carried out in accordance with the recommendations of the Declaration of Helsinki (as revised in 2013). This study was approved by Beijing Tiantan Hospital institutional review board (Approval No. KY2020-135-01). The individual consent for this retrospective analysis was waived.
Data sharing statement
The dataset used in support of the findings of this study are available from the corresponding author at [email protected] upon request.
Funding
This study received funding from the National Natural Science Foundation of China (grant number 82003192) and the Cancer Precision Radiotherapy Spark Program of the China International Medical Foundation (grant numbers 2019-N-11-37 and 2019-N-11-35).
Conflict of interest
Dr. Shibin Sun has been serving as an executive associate editor of Neurosurgical Subspecialties since July 2024. The authors report no conflicts of interest concerning the materials or methods used in this study or the findings specified in this article. Additionally, the authors declare no relationships with any companies whose products or services may be related to the subject matter of the article.
Authors’ contributions
Conceptualization (EB, KW, DG, BL, AL, SS), methodology, software, validation, form analysis, investigation, resources, data curation (EB, KW), writing - original draft (EB), writing - review & editing (KW, DG, BL, AL, SS), supervision, and project administration (SS). All authors have approved the final version and publication of the manuscript.