Lipidomic profile of meningiomas harboring different NF2 mutation status

Lipidomic profile of meningiomas harboring different NF2 mutation status | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Lipidomic profile of meningiomas harboring different NF2 mutation status Joanna Bogusiewicz, Ivana Stanimirova, Magdalena Gaca-Tabaszewska, and 9 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6490941/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 9 You are reading this latest preprint version Abstract Introduction: Meningiomas are mainly benign brain tumors, but they can evolve to higher grades. The phenomena of these changes are not well-known. Therefore, more basic research is needed. This study attempted to assess the lipidome profile in meningiomas harboring different NF2 mutation statuses (wildtype and mutated). Solid-phase microextraction (SPME) probes were used to sample and extract the metabolites and reduce the invasiveness of lipidomic analysis. Objectives: This study aimed to select the set of lipids distinguishing meningiomas with different genotypes using two chromatography methods (hydrophilic interaction chromatography (HILIC) and reversed-phase chromatography (RPLC) in two ionization modes. Methods: Brain tumors were obtained during neurosurgical procedures. Then, sampling using SPME fibers was performed directly after the lesion excision. After collecting the whole batch of samples, desorption using an isopropanol-methanol solution was performed. Subsequently, instrumental analysis was carried out using liquid chromatography coupled with high-resolution mass spectrometry. The remaining part of the lesion was stored as paraffin tissue blocks, and then genetic testing was performed to determine the presence of mutations in the NF2 gene. Results: Genetic profiling of meningiomas revealed that most lesions had a mutation in the NF2 gene. A wide range of analytes was extracted from studied tumors using SPME probes, but it was possible to select a set of 26 lipids crucial in tumor differentiation. It was also observed that a combination of analytes detected in more than one analysis mode increased the differentiation of mutant and wildtype samples, which was presented by the high sensitivity and specificity of the prepared models. Conclusions: SPME coupled liquid chromatography and mass spectrometry, can be successfully applied to the screening of lipids in meningiomas with different NF mutation statuses. meningioma brain tumor SPME lipidomics NF2 merlin Figures Figure 1 Figure 2 1. Introduction Primary brain tumors are classified by World Health Organization (WHO) recommendations (Louis et al. 2021 ). Most of them are benign lesions, but some can evolve into II or III-grade tumors (Nowosielski et al. 2017 ). Surgical removal is the best treatment option, although, in some cases, it cannot be entirely removed or is malignant, so chemotherapy or radiotherapy must be applied (Gupta et al. 2018 ; Nowosielski et al. 2017 ). Therefore, basic research in the direction enabling understanding of the relationship between genetic mutations, their translation to molecular biology, and, subsequently, the impact on mechanisms behind the sudden increase of malignancy is of great importance (Nowosielski et al. 2017 ). The WHO recommendation implies the high importance of genetic tests in diagnosing brain tumors (Louis et al. 2021 ). The most common genetic aberration in meningioma patients is a mutation in the NF2 (Lee et al. 2019 ). The protein encoded by NF2 is merlin, a member of a moesin-ezrin-radixin-like protein family, and it regulates cell adhesion, proliferation, and survival signaling and suppresses tumorigenesis (Lee et al. 2019 ). There are also mutations observed less often, such as mutations in Akt murine thymoma viral oncogene homolog 1 (ATK1), Kruppel-like factor 4 (KLF4), tumor necrosis factor receptor-associated factor 7 (TRAF7) in NF2 wildtype samples (Nowosielski et al. 2017 ). Molecular markers were proposed in diagnosis but are still not correlated with tumor type, progression, and clinical outcomes. Therefore, further research in this direction is conducted, and genetics testing is enriched with metabolite profiling (Gaca-Tabaszewska et al. 2022 ). An important group of metabolites are lipids due to their building and signaling role (Pan et al. 2021 ). Liquid chromatography coupled with mass spectrometry (LC-MS) is mainly used for analyte separation but different approaches can be applied (Cajka and Fiehn 2014 ). Reversed-phase liquid chromatography (RPLC) enables the separation of lipids based on the length of acyl chains and saturation status. Hydrophilic interaction chromatography (HILIC) and normal phase chromatography (NPLC), on the other hand, separate lipids based on lipid group affiliation (e.g., glycerolipids, glycerophospholipids, sphingolipids) (Cajka and Fiehn 2014 ). Nevertheless, apart from instrumental analysis, also sample preparation has to be conducted. Usually, collected tissue has to be homogenized, and then analytes are extracted. The most popular extraction method is liquid-liquid extraction using chloroform (Folch method) or methyl tert-butyl ether (MTBE) (Cajka and Fiehn 2014 ). These procedures are time and labor-consuming, but provide wide-range coverage of lipids. An alternative fast and easy approach is solid phase microextraction (SPME) which combines sampling and extraction in one step. Consequently, SPME offers the features unavailable for traditional methods, including extraction from intact tissue, spatial analysis, and in vivo studies, while compromising the range of extracted metabolites. This sample preparation method is referred to as a chemical biopsy due to the sampling of analytes without tissue consumption (Bogusiewicz et al. 2021 ). In brain studies, SPME was applied for the analysis of neurotransmitters in rats and monkeys, and for the pilot study of metabolomic and lipidomic profiling of the human brain in vivo (Bogusiewicz et al. 2021 ; Cudjoe et al. 2013 ; Lendor et al. 2019 ). Apart from healthy tissue, SPME was also used for brain tumor analysis (Bogusiewicz et al. 2022 , 2025 ; Goryńska et al. 2022 ) Combining SPME probe characteristics and the need for developments in meningioma diagnosis, this study aimed to assess the lipidome profile obtained using chemical biopsy in tumors with different NF2 mutation statuses. Tests using different ionization and chromatographic separation modes were used to explore the possibility of selecting the minimum number of lipids that could be successfully used to predict the genotype of meningioma. 2. Materials and methods 2.1. Chemicals and Materials Isopropanol, methanol, water, acetonitrile, ammonium acetate and acetic acid used in this research were liquid chromatography-mass spectrometry (LC-MS) grade and were purchased from Merck (Warsaw, Poland). External calibrant Pierce LTQ Velos ESI Positive Ion Calibration Solution was obtained from Thermo Scientific (San Jose, CA, USA), and fibers coated with an octadecyl (C18) were kindly provided by Supelco (Bellefonte, PA, USA). 2.2. Biological Material Brain tumors were obtained during neurosurgical procedures in the 10th Military Research Hospital and Polyclinic in Bydgoszcz. A detailed characterization of the patients included in these experiments is provided in the supplementary materials (Table S1 ). The study was approved by the Bioethical Committee in Bydgoszcz (KB 628/2015). 2.3. Histological Data and Genetic Test Results Tumor specimens were formalin-fixed and paraffin-embedded. All samples were classified by histopathological examination and graded according to WHO 2016 guidelines. The NF2 mutation was carried out using the multiplex ligation-dependent probe amplification (MLPA) method and followed the manufacturer's protocol. More details were given in the Supplementary Materials. 2.4. Chemical Biopsy (Solid-Phase Microextraction) Protocol SPME probes with 7mm C18 sorbent were used to sample the excised brain tumors. The exact protocol was described elsewhere (Bogusiewicz et al. 2022 ). Briefly, the sorbent was preconditioned in a methanol-water solution, 1:1,v/v. Then, the probe was inserted into the brain tumor for 30 minutes. Subsequently, the probe was removed from the sample, washed, and stored at -30°C until instrumental analysis. Finally, the analytes were desorbed in 150µl of isopropanol: methanol (1:1 v/v) using silanized inserts during 1h desorption under agitation at 850 rpm. Extraction blanks were also prepared (probes underwent all of SPME protocol steps, omitting extraction). 2.5. Liquid Chromatography-High Resolution Mass Spectrometry (LC-HRMS) Analysis LC-HRMS (Q Exactive Focus, Thermo Scientific, Bremen, Germany) was used for instrumental analysis. The HILIC (H) parameters were as follows: phase A—5 mM ammonium acetate in water; phase B—acetonitrile; gradient—0–2 min at 96% B, gradual decrease of B until 80% B at 15.0 min, and 15.1–21.0 min at 96% B; SeQuantZIC-cHILIC (Merck, Poznań, Poland) 3 µm 100 × 2.1 mm column; mobile phase flow rate—0.4 mL/min; oven temperature—40°C; and injection volume—10 µL. HILIC-HRMS analysis was conducted in positive (Hp) and negative (Hn) ion modes. The MS parameters were given elsewhere (Bogusiewicz et al. 2020 ). The RPLC (R) mobile phase was A: methanol: water, 40:60 with 10 mM ammonium acetate and 1 mM acetic acid, and B: isopropanol: methanol, 90:10 with 10 mM ammonium acetate and 1 mM acetic acid. Mobile phases were pumped with the flow rate: 0.2 mL/min, and the gradient was as follows: 0 min – 20% B; 1.0 min – 20% B; 1.5 min – 50% B; 7.5 min – 70% B; 13.0 min – 95% B; 17.0 min – 95% B; 17.1 min – 23.0 min – 20% B. XSelect C18 Column (Waters, Warsaw, Poland), 3.5 µm, 2.1 mm x 75 mm. The oven temperature was set at 55°C and the injection volume at 10 µL. RPLC-HRMS analysis was conducted in positive (Rp) and negative (Rn) ion modes. The MS parameters were given elsewhere (Bogusiewicz et al. 2020 ). Samples were run in full scan mode with a confirmation of inclusion list. The inclusion list was prepared based on the preliminary analysis of the pooled quality control sample run in full scan with discovery fragmentation. Identification was based on LipidSearch 4.1.30 (Thermo Fisher Scientific, San Jose, CA, USA) library search with mass accuracy of < 3ppm was searched. The MS parameters were given elsewhere (Bogusiewicz et al. 2020 ). The MS was externally calibrated every 72 h and mass accuracy was below 2ppm. Tumor samples in the sequence were randomized, and pooled quality controls (QC) were analyzed every 10–12 injections. 2.6. Data preprocessing and supervised multivariate analysis Data acquisition was performed using dedicated Thermo Scientific software: Xcalibur 4.2 (Thermo Fisher Scientific, San Jose, CA, USA). The data for the lipidomic studies was processed using LipidSearch 4.1.30 (Thermo Fisher Scientific, San Jose, CA, USA) with its accuracy set to 3 ppm and intensity threshold set to 10,000. The searched ion adducts included H + , NH 4 + , and Na + . The obtained results were then filtered using the following parameters: for extraction quality control (QC), an area coefficient of variation (CV) below 30% and not equal to 0; the QC: extraction blank area ratio above 20; and a peak quality factor above 0.85 for at least one of the studied groups. Further information on these search parameters has been detailed elsewhere (Bogusiewicz et al. 2022 ). After filtering the results, the peak areas for all detected lipids were normalized by the summary peak area of the probe, followed by autoscaling to the unitary standard deviation for each data variable (lipid). Then, a supervised multivariate analysis including the discriminant version of partial least squares (PLS-DA) regression combined with a variable selection procedure was performed for the discrimination of wildtype meningioma samples (NF2wt) from those with the NF2 mutation (NF2mt). The PLS-DA was combined with a bootstrapping procedure to estimate the quality of the models with all selected variables. The representativeness of the model set, while avoiding the possibility of including outliers, was guaranteed using the Kennard and Stone algorithm applied for each of the two groups separately with all variables (lipids). The model set was balanced, including 23 samples of each group, estimated as 75% of the less numerous group (31 NF2wt samples ). The test set included the remaining samples (8 NF2wt and 27 NF2mt samples). The lipids important for group differentiation were chosen based on the selectivity ratio (SR) value obtained as the ratio of explained variance to the residual variance for a variable (lipid) after a target projection transformation (Kvalheim 2010 ). For each (out of 1,000) bootstrap sample generated from the original model set by re-sampling with re-placement, a PLS-DA model of a definite complexity selected by leave-one-sample-out cross-validation was performed, and SRs for variables were calculated. The variables with average SR values over a given cut-off value were kept as important in the final model. The cut-off value of SR for each model was determined using the discriminating variable test (DIVA) and the SR plot (Rajalahti et al. 2009 ). DIVA is a nonparametric test that allows for the relation of the mean correct classification rate (MCCR) for variables in a given SR interval and for determining discriminatory ability in the entire SR range. The average value of the area under the receiver operating curve (AUC) was used as a figure of merit describing the model’s performance. At the same time, AUC, sensitivity, and specificity for the test set were calculated to describe the model’s predictive ability. Sensitivity is defined as the percentage of wild-type meningioma samples correctly predicted by the model. At the same time, specificity is the percentage of samples with NF2 mutation that is predicted correctly as being with NF2 mutation. A sensitivity and a specificity of 100% would be characteristic of the best model. A detailed scheme of the data analysis was presented elsewhere (Ząbek et al. 2015 ). All calculations were performed with MATLAB 2017 on a personal computer (Intel(R), Core(TM) i7-8550U CPU @ 1.80GHz, 2.00 GHz with 32GB RAM) using the Microsoft Windows 10 operating system. 3. Results Analysis of meningiomas regarding the mutation status of the NF2 gene revealed that 71% of studied brain tumors were wild-type, while 29% of them did not have this change. Subsequently, lipidomic profiling was performed using two chromatography types, RPLC and HILIC, in both ionization modes. Application of Hp enabled the detection of 186 lipid ions, which were classified as one TG, five SBP, one ST, 22 SM, seven PS, 64 PE, 46 PC, 12 LPC, 2 LPE, 12 HexCer, one Cer, and 13 acylcarnitines (Table S2). The highest number of different species of PC and PE were detected, but these phospholipids were also the most abundant analytes based on their summary peak areas. It was observed that phospholipids such as LPC, PC, PS, and PE were mainly downregulated in NF2mt samples, while plasmogens of PC and PE were upregulated (Table S2). In the SM group, selecting one trend of change was impossible (Table S2). Interestingly, it was observed that HexCer species were slightly upregulated in NF2mt tumors, but only one analyte was significantly changed (Table S2). Acylcarnitines were not significantly changed, but the trend of levels in NF2mt tumors was observed (Table S2). Next, chemometric analysis was performed. The values in Table 1 show that the model built for all lipids obtained using the Hp presents good predictive ability (AUC test =0.99). This model's sensitivity of 100% indicates the best prediction, three NF2mt samples were incorrectly predicted as NF2wt, resulting in a specificity of 88.9%. Additionaly, the PLS-DA model (AUC test =0.64) using only nine lipids selected based on SR showed a very low specificity of 37.0%, which means that ten NF2mt samples were wrongly recognized as wild-type samples. The model has a relatively good sensitivity of 75.0% (Fig. 1 , Table S6). Changing the ionization mode to negative allowed the detection of 26 lipid ions classified as 22 PE, two PC, one PS, and one PI. Similarly to previous observations, phospholipids were downregulated, but some plasmalogens were detected at higher levels in NF2mt tumors (Table S3). Compared to the positive ionization mode, the model (AUC test =0.99) for the Hn lipids showed a slightly lower sensitivity of 87.5%, but a higher specificity of 92.6%. The model with the selected five phospholipids was characterized by a lower sensitivity of 62.5% and a specificity of 81.5% in comparison to the model with all lipids (Fig. 1 , Table 1 ). A model using all lipids (212) obtained using Hp and Hn presented better prediction capability than the predictions for individual models (Table 1 ). Furthermore, the combined model based only on lipids selected based on SR also showed better sensitivity and specificity than the individual models. Table 1 The average AUC values (± uncertainty in the AUC estimation) for the model set (AUC model ) and the AUC values for the test set obtained from PLS-DA with all HILIC variables and variables selected using the SR approach. Sensitivity and specificity for the test set are also presented. Model PLS-DA (complexity AUC model AUC test Sensitivity [%] Specificity [%] Cut-off value of SR (MCCR [%]) variables Hp * 4 0.99 ± 0.01 0.99 100.0 88.9 - all (186) Hn # 2 0.94 ± 0.03 0.99 87.5 92.6 - all (26) Hp 2 0.88 ± 0.05 0.64 75.0 37.0 0.30 (60) (9) LPC 18:3, LPC 18:3, PC 35:1, PC P-39:1, PC P:36:3, PC 38:4, PS 40:7, SPB 14:0;O3, SPB 18:0;O3 Hn 1 0.80 ± 0.06 0.90 62.5 81.5 0.13 (56) (5) PE 38:1, PE P-38:6, PE O-40:6, PE P-40:6, PS 36:1 Hp + Hn 4 0.99 ± 0.01 0.99 100.0 92.6 - all (212) Hp + Hn 2 0.94 ± 0.03 0.90 75.0 77.8 0.20 (58) (14) * Hp – HILIC in positive mode # Hn - HILIC in negative mode Rp enabled the detection of 95 lipid ions, corresponding to 73 lipid species classified as one Cer, 11 HexCer, three CheE, 25 PC, 36 PE, 8 PS, 7 SM, and 3 SBP. The trend was observed that phospholipids were downregulated in NF2mt samples while plasmalogens of PC and PE were upregulated (Table S4). HexCer was downregulated in NF2mt meningiomas but only HexCer 44:2,O2 was significantly altered (p < 0.05). Sphingomyelins were altered, but there was difficulty in finding the trend of change. The PLS-DA model (AUC test =0.99) using all lipids was characterized by high sensitivity and specificity of 87.5% and 100.0%, respectively (Table 2 ). Eight lipid ions were chosen based on the SR approach and were important to distinguish between NF2wt and NF2mt samples (Table 2 , Fig. 2 ). Compared to the model with all 95 lipids, the model with the selected eight lipids (AUC test =0.90) presented a slightly lower sensitivity of 75.0% and a specificity of 88.9% (Table 2 ). Changing the ionization mode to negative allowed the detection of 42 lipid ions corresponding to 39 compounds. These analytes were classified as seven CER, six PC, 19 PE, three PS, one PI, and six SM. Similar to the observation of analytes using other modes, phospholipids with the expectation of plasmalogens were downregulated in NF2mt (Table S5). The trend of changes for SM was not obvious (Table S5). The discriminant model using all these analytes (AUC test =0.99) allowed for differentiated tumors with different NF mutation statuses with the best sensitivity of 100.0% and relatively high specificity of 92.6% (Table 2 ). The model built for the selected lipids was characterized by a sensitivity of 62.5% and a specificity of 100.0% (Fig. 2 , Table 2 ). A combination of the lipids (137) obtained in positive and negative modes gave a model with the same predictive abilities (AUCtest = 0.99) as the model for the lipids determined by Rp and only slightly worse predictions compared to the model using the lipids in the negative mode (Table 3 ). Looking at the predictive capabilities of models with fewer variables, the model (AUC test =0.97) with 13 lipids determined in Rp and Rn showed a sensitivity of 87.5% and a specificity of 96.3% (Table 2 ). Table 2 The average AUC values (± uncertainty in the AUC estimation) for the model set (AUC model ) and the AUC values for the test set obtained from PLS-DA with all RPLC variables and variables selected using SR approach. Sensitivity and specificity for the test set are also presented. Model PLS-DA (complexity AUC model AUC test Sensitivity [%] Specificity [%] Cut-off value of SR (MCCR [%]) variables Rp ^ 1 0.99 ± 0.01 0.99 87.5 100.0 - all (95) Rn + 3 0.99 ± 0.01 0.99 100.0 92.6 - all (42) Rp 2 0.79 ± 0.06 0.90 75.0 88.9 0.35 (57) (8) HexCer 43:2;O2, HexCer 43:2 + pO, HexCer44:2;O2, SM 38:5;O2, PC P-34:1, PC P-34:1, PC P-36:4, PE 38:4 Rn 1 0.85 ± 0.06 0.90 62.5 100.0 0.30 (59) (5) PE P-38:6, PE 40:5, PE P-40:7, PE O-40:7, PS 36:2 Rp + Rn 2 0.99 ± 0.01 0.99 87.5 100.0 - all (137) Rp + Rn 1 0.88 ± 0.05 0.97 87.5 96.3 0.30 (57) (13) ^ Rp – RPLC in positive mode + Rn - RPLC in negative mode From a practical point of view, using either the HILIC or RPLC for analysis is preferred. Comparing the results from both methods, the discriminant model differentiating the NF2wt meningiomas from that of NF2mt was the best in terms of predictive capabilities when using all 42 lipids determined by the Rn. An equally good model in terms of predictive abilities is the model with all 212 lipids determined by both: Hp + Hn (Table 1 ). The model using only 13 lipids selected based on SR analyzed using Rp + Rn presented a slightly worse predictive ability (Table 2 ). In the next step, combinations of lipids determined by two different chromatography methods and ionization modes in meningioma differentiation were tested. The merit data are presented in Table 3 . The best-combined models (AUC test =0.99) with the highest sensitivity and specificity of 100.0% were obtained for all 281 lipids determined by Hp + Hn + Rp, and the model using all the lipids from the two methods in both modes (Hp + Hn + Rp + Rn) (Table 3 ). There are several combined models with a sensitivity of 100.0% and a specificity of 96.3%, which means that in this case, one NF2mt meningioma sample was wrongly recognized as an NF2wt (Table 3 ). The minimal set of lipids that gave the best discrimination between NF2wt and NF2mt samples, regardless of the chromatographic method and ionization mode, were 12 lipids (Table 3 ), which were determined by a combination of Hn + Rp (Table 3 ). A similar situation but with 17 analytes was observed when Hn + Rp + Rn were combined (Table 3 ). These two sets of lipids allowed to discriminate the groups of meningioma samples with the best specificity of 100.0% and sensitivity of 87.5%. Table 3 The average AUC values (± uncertainty in the AUC estimation) for the model set (AUC model ) and the AUC values for the test set obtained from PLS-DA of various combinations HILIC and RPLC variables determined in negative and positive modes and variables reduced using the SR approach. Sensitivity and specificity for the test set are also presented. Model PLS-DA (complexity) AUC model AUC test Sensitivity [%] Specificity [%] Cut-off value of SR (MCCR [%]) Variables (number of variables) Hp * +Rp ^ 2 0.99 ± 0.01 0.99 100.0 96.3 - all (281) Hp + Rp 1 0.86 ± 0.03 0.89 75.0 85.2 0.20 (60) (17) Hp + Rn + 4 0.99 ± 0.01 0.99 100.0 92.6 - all (228) Hp + Rn 2 0.93 ± 0.03 0.88 87.5 74.1 0.20 (60) (14) Hn # +Rn 2 0.97 ± 0.02 0.99 100.0 96.3 - all (68) Hn + Rn 1 0.83 ± 0.05 0.92 87.5 85.2 0.10 (57) (15) Hn + Rp 2 0.99 ± 0.01 0.99 87.5 100.0 - all (121) Hn + Rp 1 0.89 ± 0.04 0.99 87.5 100.0 0.30 (62) (12) Hp * +Hn # +Rp ^ 2 0.99 ± 0.01 0.99 100.0 100.0 - all (281) Hp + Hn + Rp 1 0.91 ± 0.04 0.97 87.5 96.3 0.23 (59) (21) Hp + Hn + Rn + 2 0.99 ± 0.01 0.99 100.0 88.9 all (254) Hp + Hn + Rn 2 0.95 ± 0.03 0.93 87.5 77.8 0.20 (58) (18) Hn + Rp + Rn 4 0.99 ± 0.01 0.99 100.0 96.3 - all (163) Hn + Rp + Rn 1 0.91 ± 0.04 0.99 87.5 100.0 0.25 (60) (17) Hp + Hn + Rp + Rn 2 0.99 ± 0.01 0.99 100.0 100.0 - all (349) Hp + Hn + Rp + Rn 1 0.93 ± 0.04 0.99 87.5 96.3 0.20 (56) (26) * Hp – HILIC in positive mode # Hn - HILIC in negative mode ^ Rp – RPLC in positive mode + Rn - RPLC in negative mode 4. Discussion A mutation in the NF2 gene was observed in the majority of studied meningiomas, which correlates with literature reporting that this aberration can be detected in over 60% of patients (Lee et al. 2019 ). A wide range of lipids, including long-chain acylcarnitines, phospholipids, sphingolipids, and glycerides, were detected in the experiment presented herein. Previous studies on brain tumors showed a similar range of analytes (Bogusiewicz et al. 2022 ; Jarmusch et al. 2016 ; Yu et al. 2020 ). What should also be pointed out is that many PC and PE plasmalogens were extracted (Table S2, S3, S4, S5). They may constitute between 20% and 50% of the total phospholipid mass in the brain (Ferreri et al. 2023 ). However, due to high sensitivity to acids, these analytes could be easily disrupted during sample preparation. Moreover, these analytes are antioxidants, so they can potentially be engaged in redox reactions in the sampled tissue, which makes them even more unstable. Therefore, the possibility of extracting these analytes could be related to the phenomenon that enzymatic reactions are quenched by binding analytes to the sorbent, thus reducing the impact of these proteins on lipidome composition (Bogusiewicz et al. 2021 ). Consequently, SPME may be a promising sampling method that does not impact plasmalogen stability, similar to the case of oxylipins, which are unstable compounds (Napylov et al. 2020 ). Studies presented herein showed that most phospholipids were downregulated in NF2mt meningiomas. It is surprising due to the known function of merlin in the hippo pathway, which is dysregulated in cancerous cells (Xu et al. 2024 ). The hippo pathway regulates the proliferation and growth of cells, and merlin serves as an inhibitor of this pathway, inhibiting proliferation and growth. Therefore, its lack is expected to be related to cancer growth, which, in turn, is usually related to higher lipid content (Bogusiewicz et al. 2022 ). However, the increase may not have been observed due to the benign character of the studied tumors. The explanation is not clear, so this observation should be a starting point for future studies of this phenomenon. An additional observation is that plasmalogens such as PE P-38:6, PE P-40:6, PC P-34:1, and PC P-36:4 were upregulated in mutant samples and had a high impact on meningioma differentiation (Table 1 , 2 , Table S6). As mentioned earlier, these analytes participate in oxygen reactivates, mitigating damages caused by free radicals (Ferreri et al. 2023 ). Antioxidant activity could be crucial in cancer development, making these analytes one of the first potential indicators of changes (Messias et al. 2018 ). Lysophospholipids, which, among others, take part in signaling processes and cell membrane fluidity through the Lands cycle, were not significantly altered in tumors studied tumors (Hishikawa et al. 2008 ). However, chemometric models selected them as discriminatory lipids (Table 1 , 2 ). The alterations in this group of lipids were reported before, e.g. a higher content of LPC was observed in gliomas with a higher grade and IDH1/2 wildtype lesions, which both have worse clinical outcomes than low-grade and IDH1/2 mutant tumors, respectively (Bogusiewicz et al. 2022 ). It could also indicate that cancerous changes and the lack of suppression in the hippo pathway disrupt lipid turnover in cell membranes. PS should also be mentioned among phospholipids due to their role in apoptosis (Kaynak et al. 2022 ). This group is not highly altered in studied samples, but some individual species, such as PS 36:1, PS 36:2, and PS 40:7, were changed and selected as discriminatory analytes in chemometric models (Table 1 , 2 , Table S6). PS is mainly localized in the inner leaflet of the cell’s membrane in normal cells, but due to apoptosis-related changes, PS is translocated to the outer leaflet signaling phagocytic cells to engulf the apoptotic one (Furuta and Zhou 2023 ). However, in cancer cells, PS, due to the activity of flipases, is localized mainly in the outer part of the membrane, leading to immunosuppression and cancer progression (Wang et al. 2022 ). The role of PS in cancer biology is complex and observed downregulation of PS in NF2mt species can be related to a lower capability for apoptosis. Even though ceramides were not selected as discriminatory analytes, their derivates, including sphingosines, sphingomyelins, and hexylcaramides, were observed as altered in meningiomas with different NF2 statuses. Sphingolipids metabolism, where ceramide is the common substrate, can be integrated into three pathways: sphingomyelins pathway (SM are created), salvage (sphingosine is produced), and modified ceramide hydrolysis where, for instance, hexoylceramide is produced (Li et al. 2022 ). It was reported that alterations in these metabolic pathways could be related to alterations in cancer cell growth and migration, autophagy, and apoptosis processes in neoplastic lesions such as glioblastoma, breast cancer, and liver cancer (Li et al. 2022 ). In research presented herein significant alterations were observed only in sphingomyelins groups, but what is essential is that sphinganine and hexoylceramide were selected as discriminatory analytes (Table S6). It shows their impact on tumor metabolism. The common trend of sphingomyelin changes in NFmt and NFwt tumors was not evident. However, SM participates in creating lipid rafts, and their level increases corresponding to a more rigid structure (Hirano et al. 2022 ). Moreover, these analytes are the major source of ceramides, one of the most bioactive lipids (Goñi 2022 ). The study presented herein selected sphinganine and sphingosine as discriminatory lipids (Table 1 , 2 ). They are the building blocks for other sphingolipids (Farley et al. 2024 ). Thus, alterations in their level could correspond to further changes in metabolic pathways, e.g. sphinganine takes part in apoptosis, which can be induced by the cleaving of procaspase-3 (Farley et al. 2024 ). Hexyl ceramides could also be counted in sphingolipid species, and they were related to the apoptosis. Lastly, acylcarnitine levels with medium and long acyl chains were altered, although they were not selected as significantly altered or discriminatory species (Table S2). Disruption in this group of analytes was widely reported as important in tumor progression due to its correlation with tumor plasticity and the high demand for energy to grow and proliferate (Melone et al. 2018 ). Reports show that acylcarnitine levels were higher in NF2mt meningiomas than wildtype (Bogusiewicz et al. 2025 ). Interestingly the lack of merlin, protein suppressor, and microtubule stabilizer could lead to a higher possibility of developing new foci of cancer development or tumor transformation (Petrilli and Fernández-Valle 2016 ). It could explain increased energy consumption and elevated levels of lipids as markers of higher lipid turnover and increased fatty acid oxidation in NF2 mutated cells. Moreover, cancerous cell proliferation is related to higher energy consumption. This demand could be fulfilled by glucose metabolism – Warburg effect or fatty acid oxidation (Melone et al. 2018 ). Although, it was observed that NF2mt cells are characterized by higher dependence on lipid metabolism (Stepanova et al. 2017 ). Increased fatty oxidation is related to an enhanced carnitine shuttle system, which produces higher acylcarnitine levels. Finally, the selection of potential biomarkers using a particular mode of instrumental analysis resulted in worse differentiation parameters than the model's build on all detected analytes. However, combining more than one chromatography type and ionization mode increased chemometric modeling parameters. It could be related to introducing different information using different types of instrumental analysis. For instance, HILIC enables the separation of hydrophilic analytes such as phospholipids sphingomyelins, and RPLC, on the other hand, detects hydrophobic analytes such as glycerides and ceramides (Cajka and Fiehn 2014 , 2016 ). The majority of analytes are detected in positive ion mode. However, analytes such as PI, PA, PS, and fatty acids are more commonly observed in negative ionization mode (Ivanova et al. 2001 ). Moreover, adding more variables to the model allowed the preparation of a wider view of the studied tissue and increased the number of variables, which corrected the model parameters. 5. Conclusions Among lipids altered by NF2 mutation were phospholipids, sphingolipids, and acylcarnitines. Interestingly, the trend of upregulation of PC and PE plasmogenes and acylcarnitines in NF2mt meningiomas was observed. Chemometric analysis revealed that based on lipid profiles determined by both the RPLC and HPLC in positive and negative ionization modes made it possible to differentiate the NF2wt meningioma samples from those with NF2mt with the best predictive abilities with a sensitivity and specificity of 100.0%. The same prediction outcome was obtained using the whole set of lipids determined by the combination Hp + Hn and also only Rn. Two combined discriminant models using a reduced number of 12 (Hp + Rp) or 17 (Hp + Rp + Rn) lipids were also found useful, with a sensitivity of 87.5% and a specificity of 100.0%. If there is a need to limit the chromatographic methods to one, the discriminant model using all 42 lipids determined by Rn had the best sensitivity of 100.0% and a specificity of 92.6%. Abbreviations AUC – area under curve Cer – Ceramide ChE – cholesterol ester CV – coefficient of variation DIVA – discriminating variable test HexCer – hexoylcramide HILIC – Hydrophilic interaction chromatography Hn - HILIC in negative mode Hp – HILIC in positive mode KLF4 – Kruppel-like factor 4 LC-MS – Liquid chromatography coupled with mass spectrometry LPC – Lysophosphatidylcholine LPE – Lysophosphatidylethanolamine MAD – median absolute deviation MCCR – mean correct classification rate MTBE – Methyl tert-butyl ether NF2m – NF2 mutant NF2w – NF2 wildtype NPLC – normal phase chromatography PA – Phosphatidic acid PC – Phosphatidylcholine PE – Phosphatidylethanolamine PI – Phosphatidylinositol PLS-DA – partial least squares PS – Phosphatidylserine QC – quality control Rn - RPLC in negative mode ROC – receiver operating characteristic Rp – RPLC in positive mode RPLC – Reversed-phase liquid chromatography SM – Sphingomyelin SPB – Sphingosine SPME – solid-phase microextraction SR – selectivity ratio TP – target projection TRAF7 – tumor necrosis factor receptor-associated factor 7 WHO – World Health Organization Declarations Acknowledgments The National Science Centre Poland funded lipidomic profiling within research grant No. 2015/18/M/ST4/00059. The National Science Centre Poland funded the genetic tests within research grant No. 2019/33/N/ST4/00286. The authors acknowledge Supelco/MilliporeSigma for kindly supplying the SPME probes. The authors would like to acknowledge Paulina Zofia Goryńska, Krzysztof Goryński, and Karol Jaroch for their help with samplings. Conflict of interest The authors declare no conflict of interest. Author contributions Conceptualization: J.B., B.B. M.H.; Methodology: J.B.; Investigation: J.B., M.G.-T., P.S., K.S., A.M. and A.R.; Resources: M.B., B.B., J.F., and M.H.; Data curation: J.B, I. S.; Data analysis and description: I.S.; Writing, original draft preparation: J.B.; Writing, review and editing: B.B.; Visualization: J.B., I.S.; Supervision: B.B.; Project administration: J.B. and B.B.; Funding acquisition: J.B. and B.B. All authors have read and agreed to the published version of the manuscript. Ethical Statements The study was approved by the Bioethical Committee in Bydgoszcz (KB 628/2015). Data Availability The experimental data that support the findings of this study are available in the https://repod.icm.edu.pl/dataset.xhtml?persistentId=doi:10.18150/R0ZNBL References Bogusiewicz, J., Burlikowska, K., Łuczykowski, K., Jaroch, K., Birski, M., Furtak, J., et al. (2021). New chemical biopsy tool for spatially resolved profiling of human brain tissue in vivo. Scientific Reports , 11 (1), 19522. https://doi.org/10.1038/s41598-021-98973-y Bogusiewicz, J., Furtak, J., Birski, M., Soszyńska, K., Majdańska, A., Ryfa, A., et al. (2025). Acylcarnitine Profiling in Meningiomas with Different NF2 Mutation Statuses. International Journal of Molecular Sciences , 26 (4), 1570. https://doi.org/10.3390/ijms26041570 Bogusiewicz, J., Goryńska, P. Z., Gaca, M., Chmara, K., Goryński, K., Jaroch, K., et al. (2020). On-Site Sampling and Extraction of Brain Tumors for Metabolomics and Lipidomics Analysis. Journal of Visualized Experiments , 2020 (159), 159. https://doi.org/10.3791/61260 Bogusiewicz, J., Kupcewicz, B., Goryńska, P. Z., Jaroch, K., Goryński, K., Birski, M., et al. (2022). Investigating the Potential Use of Chemical Biopsy Devices to Characterize Brain Tumor Lipidomes. International Journal of Molecular Sciences , 23 (7), 3518. https://doi.org/10.3390/ijms23073518 Cajka, T., & Fiehn, O. (2014). Comprehensive analysis of lipids in biological systems by liquid chromatography-mass spectrometry. TrAC Trends in Analytical Chemistry , 61 , 192–206. https://doi.org/10.1016/j.trac.2014.04.017 Cajka, T., & Fiehn, O. (2016). Toward Merging Untargeted and Targeted Methods in Mass Spectrometry-Based Metabolomics and Lipidomics. Analytical Chemistry , 88 (1), 524–545. https://doi.org/10.1021/acs.analchem.5b04491 Cudjoe, E., Bojko, B., Delannoy, I., Saldivia, V., & Pawliszyn, J. (2013). Solid-phase microextraction: A complementary InVivo sampling method to microdialysis. Angewandte Chemie - International Edition , 52 (46), 12124–12126. https://doi.org/10.1002/anie.201304538 Farley, S., Stein, F., Haberkant, P., Tafesse, F. G., & Schultz, C. (2024). Trifunctional Sphinganine: A New Tool to Dissect Sphingolipid Function. ACS Chemical Biology , 19 (2). https://doi.org/10.1021/acschembio.3c00554 Ferreri, C., Ferocino, A., Batani, G., Chatgilialoglu, C., Randi, V., Riontino, M. V., et al. (2023). Plasmalogens: Free Radical Reactivity and Identification of Trans Isomers Relevant to Biological Membranes. Biomolecules , 13 (5). https://doi.org/10.3390/biom13050730 Furuta, Y., & Zhou, Z. (2023). How do necrotic cells expose phosphatidylserine to attract their predators—What’s unique and what’s in common with apoptotic cells. Frontiers in Cell and Developmental Biology . https://doi.org/10.3389/fcell.2023.1170551 Gaca-Tabaszewska, M., Bogusiewicz, J., & Bojko, B. (2022). Metabolomic and Lipidomic Profiling of Gliomas—A New Direction in Personalized Therapies. Cancers , 14 (20), 5041. https://doi.org/10.3390/cancers14205041 Goñi, F. M. (2022). Sphingomyelin: What is it good for? Biochemical and Biophysical Research Communications . https://doi.org/10.1016/j.bbrc.2022.08.074 Goryńska, P. Z., Chmara, K., Kupcewicz, B., Goryński, K., Jaroch, K., Paczkowski, D., et al. (2022). Metabolomic Phenotyping of Gliomas: What Can We Get with Simplified Protocol for Intact Tissue Analysis? Cancers , 14 (2), 312. https://doi.org/10.3390/cancers14020312 Gupta, S., Bi, W. L., & Dunn, I. F. (2018). Medical management of meningioma in the era of precision medicine. Neurosurgical Focus , 44 (4), E3. https://doi.org/10.3171/2018.1.FOCUS17754 Hirano, K., Kinoshita, M., & Matsumori, N. (2022). Impact of sphingomyelin acyl chain heterogeneity upon properties of raft-like membranes. Biochimica et Biophysica Acta - Biomembranes , 1864 (12). https://doi.org/10.1016/j.bbamem.2022.184036 Hishikawa, D., Shindou, H., Kobayashi, S., Nakanishi, H., Taguchi, R., & Shimizu, T. (2008). Discovery of a lysophospholipid acyltransferase family essential for membrane asymmetry and diversity. Proceedings of the National Academy of Sciences , 105 (8), 2830–2835. https://doi.org/10.1073/pnas.0712245105 Ivanova, P. T., Cerda, B. A., Horn, D. M., Cohen, J. S., McLafferty, F. W., & Brown, H. A. (2001). Electrospray ionization mass spectrometry analysis of changes in phospholipids in RBL-2H3 mastocytoma cells during degranulation. Proceedings of the National Academy of Sciences of the United States of America , 98 (13). https://doi.org/10.1073/pnas.131195098 Jarmusch, A. K., Pirro, V., Baird, Z., Hattab, E. M., Cohen-Gadol, A. A., & Cooks, R. G. (2016). Lipid and metabolite profiles of human brain tumors by desorption electrospray ionization-MS. Proceedings of the National Academy of Sciences of the United States of America , 113 (6), 1486–1491. https://doi.org/10.1073/pnas.1523306113 Kaynak, A., Davis, H. W., Kogan, A. B., Lee, J. H., Narmoneva, D. A., & Qi, X. (2022). Phosphatidylserine: The Unique Dual-Role Biomarker for Cancer Imaging and Therapy. Cancers . https://doi.org/10.3390/cancers14102536 Kvalheim, O. M. (2010). Interpretation of partial least squares regression models by means of target projection and selectivity ratio plots. Journal of Chemometrics , 24 (7–8). https://doi.org/10.1002/cem.1289 Lee, S., Karas, P. J., Hadley, C. C., Bayley V, J. C., Khan, A. B., Jalali, A., et al. (2019). The Role of Merlin/NF2 Loss in Meningioma Biology. Cancers , 11 (11), 1633. https://doi.org/10.3390/cancers11111633 Lendor, S., Hassani, S. A., Boyaci, E., Singh, V., Womelsdorf, T., & Pawliszyn, J. (2019). Solid Phase Microextraction-Based Miniaturized Probe and Protocol for Extraction of Neurotransmitters from Brains in Vivo. Analytical Chemistry , 91 (7), 4896–4905. https://doi.org/10.1021/acs.analchem.9b00995 Li, R. Z., Wang, X. R., Wang, J., Xie, C., Wang, X. X., Pan, H. D., et al. (2022). The key role of sphingolipid metabolism in cancer: New therapeutic targets, diagnostic and prognostic values, and anti-tumor immunotherapy resistance. Frontiers in Oncology . https://doi.org/10.3389/fonc.2022.941643 Louis, D. N., Perry, A., Wesseling, P., Brat, D. J., Cree, I. A., Figarella-Branger, D., et al. (2021). The 2021 WHO Classification of Tumors of the Central Nervous System: a summary. Neuro-Oncology , 23 (8), 1231–1251. https://doi.org/10.1093/neuonc/noab106 Melone, M. A. B., Valentino, A., Margarucci, S., Galderisi, U., Giordano, A., & Peluso, G. (2018). The carnitine system and cancer metabolic plasticity review-article. Cell Death and Disease , 9 , 228. https://doi.org/10.1038/s41419-018-0313-7 Messias, M. C. F., Mecatti, G. C., Priolli, D. G., & De Oliveira Carvalho, P. (2018). Plasmalogen lipids: Functional mechanism and their involvement in gastrointestinal cancer. Lipids in Health and Disease . https://doi.org/10.1186/s12944-018-0685-9 Napylov, A., Reyes-Garces, N., Gomez-Rios, G., Olkowicz, M., Lendor, S., Monnin, C., et al. (2020). In Vivo Solid-Phase Microextraction for Sampling of Oxylipins in Brain of Awake, Moving Rats. Angewandte Chemie - International Edition , 59 (6), 2392–2398. https://doi.org/10.1002/anie.201909430 Nowosielski, M., Galldiks, N., Iglseder, S., Kickingereder, P., Von Deimling, A., Bendszus, M., et al. (2017). Diagnostic challenges in meningioma. Neuro-Oncology , 19 (12). https://doi.org/10.1093/neuonc/nox101 Pan, M., Qin, C., & Han, X. (2021). Lipid Metabolism and Lipidomics Applications in Cancer Research. In Advances in Experimental Medicine and Biology (Vol. 1316, pp. 1–24). https://doi.org/10.1007/978-981-33-6785-2_1 Petrilli, A. M., & Fernández-Valle, C. (2016). Role of Merlin/NF2 inactivation in tumor biology. Oncogene , 35 (5), 537–548. https://doi.org/10.1038/onc.2015.125 Rajalahti, T., Arneberg, R., Kroksveen, A. C., Berle, M., Myhr, K. M., & Kvalheim, O. M. (2009). Discriminating variable test and selectivity ratio plot: Quantitative tools for interpretation and variable (biomarker) selection in complex spectral or chromatographic profiles. Analytical Chemistry , 81 (7). https://doi.org/10.1021/ac802514y Stepanova, D. S., Semenova, G., Kuo, Y. M., Andrews, A. J., Ammoun, S., Hanemann, C. O., & Chernoff, J. (2017). An essential role for the tumor-suppressor merlin in regulating fatty acid synthesis. Cancer Research , 77 (18). https://doi.org/10.1158/0008-5472.CAN-16-2834 Wang, W., Wu, S., Cen, Z., Zhang, Y., Chen, Y., Huang, Y., et al. (2022). Mobilizing phospholipids on tumor plasma membrane implicates phosphatidylserine externalization blockade for cancer immunotherapy. Cell Reports , 41 (5). https://doi.org/10.1016/j.celrep.2022.111582 Xu, D., Yin, S., & Shu, Y. (2024). NF2: An underestimated player in cancer metabolic reprogramming and tumor immunity. npj Precision Oncology , 8 (1), 133. https://doi.org/10.1038/s41698-024-00627-5 Yu, D., Xuan, Q., Zhang, C., Hu, C., Li, Y., Zhao, X., et al. (2020). Metabolic Alterations Related to Glioma Grading Based on Metabolomics and Lipidomics Analyses. Metabolites , 10 (12), 478. https://doi.org/10.3390/metabo10120478 Ząbek, A., Stanimirova, I., Deja, S., Barg, W., Kowal, A., Korzeniewska, A., et al. (2015). Fusion of the 1H NMR data of serum, urine and exhaled breath condensate in order to discriminate chronic obstructive pulmonary disease and obstructive sleep apnea syndrome. Metabolomics , 11 (6). https://doi.org/10.1007/s11306-015-0808-5 Additional Declarations No competing interests reported. Supplementary Files BogusiewiczetalSuppl.docx floatimage1.jpeg Graphical abstract Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 06 Aug, 2025 Reviews received at journal 03 Aug, 2025 Reviews received at journal 10 Jul, 2025 Reviewers agreed at journal 12 Jun, 2025 Reviewers agreed at journal 01 May, 2025 Reviewers invited by journal 28 Apr, 2025 Editor assigned by journal 20 Apr, 2025 Submission checks completed at journal 20 Apr, 2025 First submitted to journal 20 Apr, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6490941","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":449329499,"identity":"9a9d6be2-cef9-4850-a990-2d95df878814","order_by":0,"name":"Joanna Bogusiewicz","email":"","orcid":"","institution":"Nicolaus Copernicus University","correspondingAuthor":false,"prefix":"","firstName":"Joanna","middleName":"","lastName":"Bogusiewicz","suffix":""},{"id":449329501,"identity":"89c835fc-18fd-4387-b70e-5960ee1e1e54","order_by":1,"name":"Ivana Stanimirova","email":"","orcid":"","institution":"University of Silesia in Katowice","correspondingAuthor":false,"prefix":"","firstName":"Ivana","middleName":"","lastName":"Stanimirova","suffix":""},{"id":449329503,"identity":"a6a2a290-15b8-476b-85a4-eb7428ec2929","order_by":2,"name":"Magdalena Gaca-Tabaszewska","email":"","orcid":"","institution":"Nicolaus Copernicus University","correspondingAuthor":false,"prefix":"","firstName":"Magdalena","middleName":"","lastName":"Gaca-Tabaszewska","suffix":""},{"id":449329504,"identity":"373f08fa-c575-4060-a368-67ba8b5f8f68","order_by":3,"name":"Paulina Szeliska","email":"","orcid":"","institution":"Nicolaus Copernicus University","correspondingAuthor":false,"prefix":"","firstName":"Paulina","middleName":"","lastName":"Szeliska","suffix":""},{"id":449329506,"identity":"7fd1bbff-7fcb-455b-b29b-2addad7cd7ce","order_by":4,"name":"Krystyna Soszyńska","email":"","orcid":"","institution":"10th Military Research Hospital and Polyclinic","correspondingAuthor":false,"prefix":"","firstName":"Krystyna","middleName":"","lastName":"Soszyńska","suffix":""},{"id":449329508,"identity":"d4c9b8ac-94a8-476d-a993-926e83630a20","order_by":5,"name":"Anna Majdańska","email":"","orcid":"","institution":"10th Military Research Hospital and Polyclinic","correspondingAuthor":false,"prefix":"","firstName":"Anna","middleName":"","lastName":"Majdańska","suffix":""},{"id":449329509,"identity":"be3e2053-b070-4944-84c1-85b5b28d6033","order_by":6,"name":"Agata Ryfa","email":"","orcid":"","institution":"10th Military Research Hospital and Polyclinic","correspondingAuthor":false,"prefix":"","firstName":"Agata","middleName":"","lastName":"Ryfa","suffix":""},{"id":449329512,"identity":"95129b95-0305-4743-8e60-c46b222dcede","order_by":7,"name":"Alicja Bartoszewska-Kubiak","email":"","orcid":"","institution":"10th Military Research Hospital and Polyclinic","correspondingAuthor":false,"prefix":"","firstName":"Alicja","middleName":"","lastName":"Bartoszewska-Kubiak","suffix":""},{"id":449329513,"identity":"a7b8512a-d37e-4d6e-a1d9-2cb4d21d820e","order_by":8,"name":"Jacek Furtak","email":"","orcid":"","institution":"Bydgoszcz University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Jacek","middleName":"","lastName":"Furtak","suffix":""},{"id":449329514,"identity":"a4e9179a-0c46-4441-bf2d-1d911d50f078","order_by":9,"name":"Marcin Birski","email":"","orcid":"","institution":"10th Military Research Hospital and Polyclinic","correspondingAuthor":false,"prefix":"","firstName":"Marcin","middleName":"","lastName":"Birski","suffix":""},{"id":449329515,"identity":"818c262d-637b-4f33-a50d-a1aa728a1647","order_by":10,"name":"Marek Harat","email":"","orcid":"","institution":"Bydgoszcz University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Marek","middleName":"","lastName":"Harat","suffix":""},{"id":449329516,"identity":"b85d2bd4-a038-416a-8158-3d6640e176db","order_by":11,"name":"Barbara Bojko","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAs0lEQVRIiWNgGAWjYBACAwkGBsMGBhswh7GBgUGGWC1pcC08RGkBqjxMghZz6eYDhTMqzsvrNvCYbpzBcIewFss5xxIMN5y5bbjtAI/ZzQ0Mz4hw2I0cA8OHbbcTzEBaHjAcJlrLOVK1bGw7ANGygRgtYL/MOJNsuO0wW9nNGQZE+AUYYscMeyrs5M2ON2+72VNxR46gFiBgMwBTzGB3HiBGBwPzAyQOcVpGwSgYBaNgZAEA2xFBsQAIPzMAAAAASUVORK5CYII=","orcid":"","institution":"Nicolaus Copernicus University","correspondingAuthor":true,"prefix":"","firstName":"Barbara","middleName":"","lastName":"Bojko","suffix":""}],"badges":[],"createdAt":"2025-04-20 20:08:06","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6490941/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6490941/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":81656193,"identity":"99328d7c-faa0-4e90-95f0-b06ccf8f1a3d","added_by":"auto","created_at":"2025-04-29 18:21:58","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":246361,"visible":true,"origin":"","legend":"\u003cp\u003ePLS-DA with TP for metabolites obtained from the Hp and Hn: a) TP scores vector from the model with all Hp metabolites, b) TP loadings vector with the selected Hp metabolites (Table 1), c) TP scores vector from the model with Hp selected metabolites by SR, d) TP scores vector from the model with all Hn metabolites, e) TP loadings vector with the selected Hn metabolites (Table 1), f) TP scores vector from the model with Hn selected metabolites by SR, g) TP scores vector from the model with all Hp+Hn metabolites, h) TP loadings vector with the selected Hp+Hn metabolites (Table 1), i) TP scores vector from the model with Hp+Hn selected metabolites by SR.\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-6490941/v1/d1744f2b8fef99ca6487be8c.png"},{"id":81656192,"identity":"09464f7c-27ef-48ce-92f9-35a69bf7b41c","added_by":"auto","created_at":"2025-04-29 18:21:58","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":234888,"visible":true,"origin":"","legend":"\u003cp\u003ePLS-DA with TP for metabolites obtained from the Rp and Rn: a) TP scores vector from the model with all Rp metabolites, b) TP loadings vector with the selected Rp metabolites (Table 2), c) TP scores vector from the model with Rp selected metabolites by SR), d) TP scores vector from the model with all Rn metabolites, e) TP loadings vector with the selected Rn metabolites (Table 2), f) TP scores vector from the model with Rn selected metabolites by SR, g) TP scores vector from the model with all Rp+Rn metabolites, h) TP loadings vector with the selected Rp+Rn metabolites (Table 2), i) TP scores vector from the model with Rp+Rn selected metabolites by SR.\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-6490941/v1/d57b4edfe5f96d54a1776aed.png"},{"id":81657338,"identity":"d2f9f486-8b2c-489b-af72-1e9b0792c1e9","added_by":"auto","created_at":"2025-04-29 18:53:58","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1481385,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6490941/v1/391137ba-393a-4d78-a510-66beb50182a3.pdf"},{"id":81656792,"identity":"2bfade2e-3aff-4f19-9571-6db0a8cbf1fb","added_by":"auto","created_at":"2025-04-29 18:37:58","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":126327,"visible":true,"origin":"","legend":"","description":"","filename":"BogusiewiczetalSuppl.docx","url":"https://assets-eu.researchsquare.com/files/rs-6490941/v1/92dfd2bff283c3ad34d0851b.docx"},{"id":81656195,"identity":"0eee2926-77f7-46ae-941a-217d5a05b82b","added_by":"auto","created_at":"2025-04-29 18:21:58","extension":"jpeg","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":305742,"visible":true,"origin":"","legend":"\u003cp\u003eGraphical abstract\u003c/p\u003e","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-6490941/v1/1da6530f66546ef3ce968fbe.jpeg"}],"financialInterests":"No competing interests reported.","formattedTitle":"Lipidomic profile of meningiomas harboring different NF2 mutation status","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003ePrimary brain tumors are classified by World Health Organization (WHO) recommendations (Louis et al. \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Most of them are benign lesions, but some can evolve into II or III-grade tumors (Nowosielski et al. \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). Surgical removal is the best treatment option, although, in some cases, it cannot be entirely removed or is malignant, so chemotherapy or radiotherapy must be applied (Gupta et al. \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Nowosielski et al. \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). Therefore, basic research in the direction enabling understanding of the relationship between genetic mutations, their translation to molecular biology, and, subsequently, the impact on mechanisms behind the sudden increase of malignancy is of great importance (Nowosielski et al. \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2017\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe WHO recommendation implies the high importance of genetic tests in diagnosing brain tumors (Louis et al. \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). The most common genetic aberration in meningioma patients is a mutation in the NF2 (Lee et al. \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). The protein encoded by NF2 is merlin, a member of a moesin-ezrin-radixin-like protein family, and it regulates cell adhesion, proliferation, and survival signaling and suppresses tumorigenesis (Lee et al. \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). There are also mutations observed less often, such as mutations in Akt murine thymoma viral oncogene homolog 1 (ATK1), Kruppel-like factor 4 (KLF4), tumor necrosis factor receptor-associated factor 7 (TRAF7) in NF2 wildtype samples (Nowosielski et al. \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2017\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eMolecular markers were proposed in diagnosis but are still not correlated with tumor type, progression, and clinical outcomes. Therefore, further research in this direction is conducted, and genetics testing is enriched with metabolite profiling (Gaca-Tabaszewska et al. \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). An important group of metabolites are lipids due to their building and signaling role (Pan et al. \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2021\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eLiquid chromatography coupled with mass spectrometry (LC-MS) is mainly used for analyte separation but different approaches can be applied (Cajka and Fiehn \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2014\u003c/span\u003e). Reversed-phase liquid chromatography (RPLC) enables the separation of lipids based on the length of acyl chains and saturation status. Hydrophilic interaction chromatography (HILIC) and normal phase chromatography (NPLC), on the other hand, separate lipids based on lipid group affiliation (e.g., glycerolipids, glycerophospholipids, sphingolipids) (Cajka and Fiehn \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2014\u003c/span\u003e). Nevertheless, apart from instrumental analysis, also sample preparation has to be conducted. Usually, collected tissue has to be homogenized, and then analytes are extracted. The most popular extraction method is liquid-liquid extraction using chloroform (Folch method) or methyl tert-butyl ether (MTBE) (Cajka and Fiehn \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2014\u003c/span\u003e). These procedures are time and labor-consuming, but provide wide-range coverage of lipids. An alternative fast and easy approach is solid phase microextraction (SPME) which combines sampling and extraction in one step. Consequently, SPME offers the features unavailable for traditional methods, including extraction from intact tissue, spatial analysis, and \u003cem\u003ein vivo\u003c/em\u003e studies, while compromising the range of extracted metabolites. This sample preparation method is referred to as a chemical biopsy due to the sampling of analytes without tissue consumption (Bogusiewicz et al. \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). In brain studies, SPME was applied for the analysis of neurotransmitters in rats and monkeys, and for the pilot study of metabolomic and lipidomic profiling of the human brain \u003cem\u003ein vivo\u003c/em\u003e (Bogusiewicz et al. \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Cudjoe et al. \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2013\u003c/span\u003e; Lendor et al. \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). Apart from healthy tissue, SPME was also used for brain tumor analysis (Bogusiewicz et al. \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2022\u003c/span\u003e, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Goryńska et al. \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2022\u003c/span\u003e)\u003c/p\u003e \u003cp\u003eCombining SPME probe characteristics and the need for developments in meningioma diagnosis, this study aimed to assess the lipidome profile obtained using chemical biopsy in tumors with different NF2 mutation statuses. Tests using different ionization and chromatographic separation modes were used to explore the possibility of selecting the minimum number of lipids that could be successfully used to predict the genotype of meningioma.\u003c/p\u003e"},{"header":"2. Materials and methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1. Chemicals and Materials\u003c/h2\u003e \u003cp\u003eIsopropanol, methanol, water, acetonitrile, ammonium acetate and acetic acid used in this research were liquid chromatography-mass spectrometry (LC-MS) grade and were purchased from Merck (Warsaw, Poland). External calibrant Pierce LTQ Velos ESI Positive Ion Calibration Solution was obtained from Thermo Scientific (San Jose, CA, USA), and fibers coated with an octadecyl (C18) were kindly provided by Supelco (Bellefonte, PA, USA).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e2.2. Biological Material\u003c/h2\u003e \u003cp\u003eBrain tumors were obtained during neurosurgical procedures in the 10th Military Research Hospital and Polyclinic in Bydgoszcz. A detailed characterization of the patients included in these experiments is provided in the supplementary materials (Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e). The study was approved by the Bioethical Committee in Bydgoszcz (KB 628/2015).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e2.3. Histological Data and Genetic Test Results\u003c/h2\u003e \u003cp\u003eTumor specimens were formalin-fixed and paraffin-embedded. All samples were classified by histopathological examination and graded according to WHO 2016 guidelines. The NF2 mutation was carried out using the multiplex ligation-dependent probe amplification (MLPA) method and followed the manufacturer's protocol. More details were given in the Supplementary Materials.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003e2.4. Chemical Biopsy (Solid-Phase Microextraction) Protocol\u003c/h2\u003e \u003cp\u003eSPME probes with 7mm C18 sorbent were used to sample the excised brain tumors. The exact protocol was described elsewhere (Bogusiewicz et al. \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Briefly, the sorbent was preconditioned in a methanol-water solution, 1:1,v/v. Then, the probe was inserted into the brain tumor for 30 minutes. Subsequently, the probe was removed from the sample, washed, and stored at -30\u0026deg;C until instrumental analysis. Finally, the analytes were desorbed in 150\u0026micro;l of isopropanol: methanol (1:1 v/v) using silanized inserts during 1h desorption under agitation at 850 rpm. Extraction blanks were also prepared (probes underwent all of SPME protocol steps, omitting extraction).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003e2.5. Liquid Chromatography-High Resolution Mass Spectrometry (LC-HRMS) Analysis\u003c/h2\u003e \u003cp\u003eLC-HRMS (Q Exactive Focus, Thermo Scientific, Bremen, Germany) was used for instrumental analysis.\u003c/p\u003e \u003cp\u003eThe HILIC (H) parameters were as follows: phase A\u0026mdash;5 mM ammonium acetate in water; phase B\u0026mdash;acetonitrile; gradient\u0026mdash;0\u0026ndash;2 min at 96% B, gradual decrease of B until 80% B at 15.0 min, and 15.1\u0026ndash;21.0 min at 96% B; SeQuantZIC-cHILIC (Merck, Poznań, Poland) 3 \u0026micro;m 100 \u0026times; 2.1 mm column; mobile phase flow rate\u0026mdash;0.4 mL/min; oven temperature\u0026mdash;40\u0026deg;C; and injection volume\u0026mdash;10 \u0026micro;L. HILIC-HRMS analysis was conducted in positive (Hp) and negative (Hn) ion modes. The MS parameters were given elsewhere (Bogusiewicz et al. \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe RPLC (R) mobile phase was A: methanol: water, 40:60 with 10 mM ammonium acetate and 1 mM acetic acid, and B: isopropanol: methanol, 90:10 with 10 mM ammonium acetate and 1 mM acetic acid. Mobile phases were pumped with the flow rate: 0.2 mL/min, and the gradient was as follows: 0 min \u0026ndash; 20% B; 1.0 min \u0026ndash; 20% B; 1.5 min \u0026ndash; 50% B; 7.5 min \u0026ndash; 70% B; 13.0 min \u0026ndash; 95% B; 17.0 min \u0026ndash; 95% B; 17.1 min \u0026ndash; 23.0 min \u0026ndash; 20% B. XSelect C18 Column (Waters, Warsaw, Poland), 3.5 \u0026micro;m, 2.1 mm x 75 mm. The oven temperature was set at 55\u0026deg;C and the injection volume at 10 \u0026micro;L. RPLC-HRMS analysis was conducted in positive (Rp) and negative (Rn) ion modes. The MS parameters were given elsewhere (Bogusiewicz et al. \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eSamples were run in full scan mode with a confirmation of inclusion list. The inclusion list was prepared based on the preliminary analysis of the pooled quality control sample run in full scan with discovery fragmentation. Identification was based on LipidSearch 4.1.30 (Thermo Fisher Scientific, San Jose, CA, USA) library search with mass accuracy of \u0026lt;\u0026thinsp;3ppm was searched. The MS parameters were given elsewhere (Bogusiewicz et al. \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe MS was externally calibrated every 72 h and mass accuracy was below 2ppm. Tumor samples in the sequence were randomized, and pooled quality controls (QC) were analyzed every 10\u0026ndash;12 injections.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003e2.6. Data preprocessing and supervised multivariate analysis\u003c/h2\u003e \u003cp\u003eData acquisition was performed using dedicated Thermo Scientific software: Xcalibur 4.2 (Thermo Fisher Scientific, San Jose, CA, USA). The data for the lipidomic studies was processed using LipidSearch 4.1.30 (Thermo Fisher Scientific, San Jose, CA, USA) with its accuracy set to 3 ppm and intensity threshold set to 10,000. The searched ion adducts included H\u003csup\u003e+\u003c/sup\u003e, NH\u003csub\u003e4\u003c/sub\u003e\u003csup\u003e+\u003c/sup\u003e, and Na\u003csup\u003e+\u003c/sup\u003e. The obtained results were then filtered using the following parameters: for extraction quality control (QC), an area coefficient of variation (CV) below 30% and not equal to 0; the QC: extraction blank area ratio above 20; and a peak quality factor above 0.85 for at least one of the studied groups. Further information on these search parameters has been detailed elsewhere (Bogusiewicz et al. \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). After filtering the results, the peak areas for all detected lipids were normalized by the summary peak area of the probe, followed by autoscaling to the unitary standard deviation for each data variable (lipid). Then, a supervised multivariate analysis including the discriminant version of partial least squares (PLS-DA) regression combined with a variable selection procedure was performed for the discrimination of wildtype meningioma samples (NF2wt) from those with the NF2 mutation (NF2mt).\u003c/p\u003e \u003cp\u003eThe PLS-DA was combined with a bootstrapping procedure to estimate the quality of the models with all selected variables. The representativeness of the model set, while avoiding the possibility of including outliers, was guaranteed using the Kennard and Stone algorithm applied for each of the two groups separately with all variables (lipids). The model set was balanced, including 23 samples of each group, estimated as 75% of the less numerous group (31 NF2wt samples ). The test set included the remaining samples (8 NF2wt and 27 NF2mt samples). The lipids important for group differentiation were chosen based on the selectivity ratio (SR) value obtained as the ratio of explained variance to the residual variance for a variable (lipid) after a target projection transformation (Kvalheim \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2010\u003c/span\u003e). For each (out of 1,000) bootstrap sample generated from the original model set by re-sampling with re-placement, a PLS-DA model of a definite complexity selected by leave-one-sample-out cross-validation was performed, and SRs for variables were calculated. The variables with average SR values over a given cut-off value were kept as important in the final model. The cut-off value of SR for each model was determined using the discriminating variable test (DIVA) and the SR plot (Rajalahti et al. \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2009\u003c/span\u003e). DIVA is a nonparametric test that allows for the relation of the mean correct classification rate (MCCR) for variables in a given SR interval and for determining discriminatory ability in the entire SR range. The average value of the area under the receiver operating curve (AUC) was used as a figure of merit describing the model\u0026rsquo;s performance. At the same time, AUC, sensitivity, and specificity for the test set were calculated to describe the model\u0026rsquo;s predictive ability. Sensitivity is defined as the percentage of wild-type meningioma samples correctly predicted by the model. At the same time, specificity is the percentage of samples with NF2 mutation that is predicted correctly as being with NF2 mutation. A sensitivity and a specificity of 100% would be characteristic of the best model. A detailed scheme of the data analysis was presented elsewhere (Ząbek et al. \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2015\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eAll calculations were performed with MATLAB 2017 on a personal computer (Intel(R), Core(TM) i7-8550U CPU @ 1.80GHz, 2.00 GHz with 32GB RAM) using the Microsoft Windows 10 operating system.\u003c/p\u003e \u003c/div\u003e"},{"header":"3. Results","content":"\u003cp\u003eAnalysis of meningiomas regarding the mutation status of the NF2 gene revealed that 71% of studied brain tumors were wild-type, while 29% of them did not have this change. Subsequently, lipidomic profiling was performed using two chromatography types, RPLC and HILIC, in both ionization modes.\u003c/p\u003e \u003cp\u003eApplication of Hp enabled the detection of 186 lipid ions, which were classified as one TG, five SBP, one ST, 22 SM, seven PS, 64 PE, 46 PC, 12 LPC, 2 LPE, 12 HexCer, one Cer, and 13 acylcarnitines (Table S2). The highest number of different species of PC and PE were detected, but these phospholipids were also the most abundant analytes based on their summary peak areas. It was observed that phospholipids such as LPC, PC, PS, and PE were mainly downregulated in NF2mt samples, while plasmogens of PC and PE were upregulated (Table S2). In the SM group, selecting one trend of change was impossible (Table S2). Interestingly, it was observed that HexCer species were slightly upregulated in NF2mt tumors, but only one analyte was significantly changed (Table S2). Acylcarnitines were not significantly changed, but the trend of levels in NF2mt tumors was observed (Table S2).\u003c/p\u003e \u003cp\u003eNext, chemometric analysis was performed. The values in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e show that the model built for all lipids obtained using the Hp presents good predictive ability (AUC\u003csub\u003etest\u003c/sub\u003e=0.99). This model's sensitivity of 100% indicates the best prediction, three NF2mt samples were incorrectly predicted as NF2wt, resulting in a specificity of 88.9%. Additionaly, the PLS-DA model (AUC\u003csub\u003etest\u003c/sub\u003e=0.64) using only nine lipids selected based on SR showed a very low specificity of 37.0%, which means that ten NF2mt samples were wrongly recognized as wild-type samples. The model has a relatively good sensitivity of 75.0% (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, Table S6).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eChanging the ionization mode to negative allowed the detection of 26 lipid ions classified as 22 PE, two PC, one PS, and one PI. Similarly to previous observations, phospholipids were downregulated, but some plasmalogens were detected at higher levels in NF2mt tumors (Table S3). Compared to the positive ionization mode, the model (AUC\u003csub\u003etest\u003c/sub\u003e=0.99) for the Hn lipids showed a slightly lower sensitivity of 87.5%, but a higher specificity of 92.6%. The model with the selected five phospholipids was characterized by a lower sensitivity of 62.5% and a specificity of 81.5% in comparison to the model with all lipids (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eA model using all lipids (212) obtained using Hp and Hn presented better prediction capability than the predictions for individual models (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Furthermore, the combined model based only on lipids selected based on SR also showed better sensitivity and specificity than the individual models.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eThe average AUC values (\u0026plusmn;\u0026thinsp;uncertainty in the AUC estimation) for the model set (AUC\u003csub\u003emodel\u003c/sub\u003e) and the AUC values for the test set obtained from PLS-DA with all HILIC variables and variables selected using the SR approach. Sensitivity and specificity for the test set are also presented.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"8\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\"\u0026plusmn;\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eModel\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePLS-DA\u003c/p\u003e \u003cp\u003e(complexity\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAUC\u003csub\u003emodel\u003c/sub\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAUC\u003csub\u003etest\u003c/sub\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSensitivity [%]\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eSpecificity\u003c/p\u003e \u003cp\u003e[%]\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eCut-off value of SR (MCCR [%])\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003evariables\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHp\u003csup\u003e*\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e \u003cp\u003e0.99\u0026thinsp;\u0026plusmn;\u0026thinsp;0.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.99\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e100.0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e88.9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eall (186)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHn\u003csup\u003e#\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e \u003cp\u003e0.94\u0026thinsp;\u0026plusmn;\u0026thinsp;0.03\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.99\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e87.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e92.6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eall (26)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHp\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e \u003cp\u003e0.88\u0026thinsp;\u0026plusmn;\u0026thinsp;0.05\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.64\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e75.0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e37.0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.30 (60)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e(9)\u003c/p\u003e \u003cp\u003eLPC 18:3, LPC 18:3, PC 35:1, PC P-39:1, PC P:36:3, PC 38:4, PS 40:7, SPB 14:0;O3, SPB 18:0;O3\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHn\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e \u003cp\u003e0.80\u0026thinsp;\u0026plusmn;\u0026thinsp;0.06\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.90\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e62.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e81.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.13 (56)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e(5)\u003c/p\u003e \u003cp\u003ePE 38:1, PE P-38:6, PE O-40:6, PE P-40:6, PS 36:1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHp\u0026thinsp;+\u0026thinsp;Hn\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e \u003cp\u003e0.99\u0026thinsp;\u0026plusmn;\u0026thinsp;0.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.99\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e100.0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e92.6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eall (212)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHp\u0026thinsp;+\u0026thinsp;Hn\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e \u003cp\u003e0.94\u0026thinsp;\u0026plusmn;\u0026thinsp;0.03\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.90\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e75.0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e77.8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.20 (58)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e(14)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"8\"\u003e\u003csup\u003e*\u003c/sup\u003eHp \u0026ndash; HILIC in positive mode\u003c/td\u003e\u003c/tr\u003e \u003ctr\u003e\u003ctd colspan=\"8\"\u003e\u003csup\u003e#\u003c/sup\u003eHn - HILIC in negative mode\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eRp enabled the detection of 95 lipid ions, corresponding to 73 lipid species classified as one Cer, 11 HexCer, three CheE, 25 PC, 36 PE, 8 PS, 7 SM, and 3 SBP. The trend was observed that phospholipids were downregulated in NF2mt samples while plasmalogens of PC and PE were upregulated (Table S4). HexCer was downregulated in NF2mt meningiomas but only HexCer 44:2,O2 was significantly altered (p\u0026thinsp;\u0026lt;\u0026thinsp;0.05). Sphingomyelins were altered, but there was difficulty in finding the trend of change.\u003c/p\u003e \u003cp\u003eThe PLS-DA model (AUC\u003csub\u003etest\u003c/sub\u003e=0.99) using all lipids was characterized by high sensitivity and specificity of 87.5% and 100.0%, respectively (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). Eight lipid ions were chosen based on the SR approach and were important to distinguish between NF2wt and NF2mt samples (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e, Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e2\u003c/span\u003e). Compared to the model with all 95 lipids, the model with the selected eight lipids (AUC\u003csub\u003etest\u003c/sub\u003e=0.90) presented a slightly lower sensitivity of 75.0% and a specificity of 88.9% (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eChanging the ionization mode to negative allowed the detection of 42 lipid ions corresponding to 39 compounds. These analytes were classified as seven CER, six PC, 19 PE, three PS, one PI, and six SM. Similar to the observation of analytes using other modes, phospholipids with the expectation of plasmalogens were downregulated in NF2mt (Table S5). The trend of changes for SM was not obvious (Table S5).\u003c/p\u003e \u003cp\u003eThe discriminant model using all these analytes (AUC\u003csub\u003etest\u003c/sub\u003e=0.99) allowed for differentiated tumors with different NF mutation statuses with the best sensitivity of 100.0% and relatively high specificity of 92.6% (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). The model built for the selected lipids was characterized by a sensitivity of 62.5% and a specificity of 100.0% (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e2\u003c/span\u003e, Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eA combination of the lipids (137) obtained in positive and negative modes gave a model with the same predictive abilities (AUCtest\u0026thinsp;=\u0026thinsp;0.99) as the model for the lipids determined by Rp and only slightly worse predictions compared to the model using the lipids in the negative mode (Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). Looking at the predictive capabilities of models with fewer variables, the model (AUC\u003csub\u003etest\u003c/sub\u003e=0.97) with 13 lipids determined in Rp and Rn showed a sensitivity of 87.5% and a specificity of 96.3% (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eThe average AUC values (\u0026plusmn;\u0026thinsp;uncertainty in the AUC estimation) for the model set (AUC\u003csub\u003emodel\u003c/sub\u003e) and the AUC values for the test set obtained from PLS-DA with all RPLC variables and variables selected using SR approach. Sensitivity and specificity for the test set are also presented.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"8\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\"\u0026plusmn;\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eModel\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePLS-DA\u003c/p\u003e \u003cp\u003e(complexity\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAUC\u003csub\u003emodel\u003c/sub\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAUC\u003csub\u003etest\u003c/sub\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSensitivity [%]\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eSpecificity\u003c/p\u003e \u003cp\u003e[%]\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eCut-off value of SR (MCCR [%])\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003evariables\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRp\u003csup\u003e^\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e \u003cp\u003e0.99\u0026thinsp;\u0026plusmn;\u0026thinsp;0.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.99\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e87.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e100.0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eall (95)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRn\u003csup\u003e+\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e \u003cp\u003e0.99\u0026thinsp;\u0026plusmn;\u0026thinsp;0.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.99\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e100.0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e92.6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eall (42)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRp\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e \u003cp\u003e0.79\u0026thinsp;\u0026plusmn;\u0026thinsp;0.06\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.90\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e75.0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e88.9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.35 (57)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e(8)\u003c/p\u003e \u003cp\u003eHexCer 43:2;O2, HexCer 43:2\u0026thinsp;+\u0026thinsp;pO, HexCer44:2;O2, SM 38:5;O2, PC P-34:1, PC P-34:1, PC P-36:4, PE 38:4\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRn\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e \u003cp\u003e0.85\u0026thinsp;\u0026plusmn;\u0026thinsp;0.06\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.90\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e62.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e100.0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.30 (59)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e(5)\u003c/p\u003e \u003cp\u003ePE P-38:6, PE 40:5, PE P-40:7, PE O-40:7, PS 36:2\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRp\u0026thinsp;+\u0026thinsp;Rn\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e \u003cp\u003e0.99\u0026thinsp;\u0026plusmn;\u0026thinsp;0.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.99\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e87.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e100.0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eall (137)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRp\u0026thinsp;+\u0026thinsp;Rn\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e \u003cp\u003e0.88\u0026thinsp;\u0026plusmn;\u0026thinsp;0.05\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.97\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e87.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e96.3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.30 (57)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e(13)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"8\"\u003e\u003csup\u003e^\u003c/sup\u003eRp \u0026ndash; RPLC in positive mode\u003c/td\u003e\u003c/tr\u003e \u003ctr\u003e\u003ctd colspan=\"8\"\u003e\u003csup\u003e+\u003c/sup\u003eRn - RPLC in negative mode\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eFrom a practical point of view, using either the HILIC or RPLC for analysis is preferred. Comparing the results from both methods, the discriminant model differentiating the NF2wt meningiomas from that of NF2mt was the best in terms of predictive capabilities when using all 42 lipids determined by the Rn. An equally good model in terms of predictive abilities is the model with all 212 lipids determined by both: Hp\u0026thinsp;+\u0026thinsp;Hn (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). The model using only 13 lipids selected based on SR analyzed using Rp\u0026thinsp;+\u0026thinsp;Rn presented a slightly worse predictive ability (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eIn the next step, combinations of lipids determined by two different chromatography methods and ionization modes in meningioma differentiation were tested. The merit data are presented in Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e. The best-combined models (AUC\u003csub\u003etest\u003c/sub\u003e=0.99) with the highest sensitivity and specificity of 100.0% were obtained for all 281 lipids determined by Hp\u0026thinsp;+\u0026thinsp;Hn\u0026thinsp;+\u0026thinsp;Rp, and the model using all the lipids from the two methods in both modes (Hp\u0026thinsp;+\u0026thinsp;Hn\u0026thinsp;+\u0026thinsp;Rp\u0026thinsp;+\u0026thinsp;Rn) (Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). There are several combined models with a sensitivity of 100.0% and a specificity of 96.3%, which means that in this case, one NF2mt meningioma sample was wrongly recognized as an NF2wt (Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe minimal set of lipids that gave the best discrimination between NF2wt and NF2mt samples, regardless of the chromatographic method and ionization mode, were 12 lipids (Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e), which were determined by a combination of Hn\u0026thinsp;+\u0026thinsp;Rp (Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). A similar situation but with 17 analytes was observed when Hn\u0026thinsp;+\u0026thinsp;Rp\u0026thinsp;+\u0026thinsp;Rn were combined (Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). These two sets of lipids allowed to discriminate the groups of meningioma samples with the best specificity of 100.0% and sensitivity of 87.5%.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eThe average AUC values (\u0026plusmn;\u0026thinsp;uncertainty in the AUC estimation) for the model set (AUC\u003csub\u003emodel\u003c/sub\u003e) and the AUC values for the test set obtained from PLS-DA of various combinations HILIC and RPLC variables determined in negative and positive modes and variables reduced using the SR approach. Sensitivity and specificity for the test set are also presented.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"8\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\"\u0026plusmn;\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eModel\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePLS-DA\u003c/p\u003e \u003cp\u003e(complexity)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAUC\u003csub\u003emodel\u003c/sub\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAUC\u003csub\u003etest\u003c/sub\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSensitivity [%]\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eSpecificity\u003c/p\u003e \u003cp\u003e[%]\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eCut-off value of SR (MCCR [%])\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eVariables\u003c/p\u003e \u003cp\u003e(number of variables)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHp\u003csup\u003e*\u003c/sup\u003e+Rp\u003csup\u003e^\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e \u003cp\u003e0.99\u0026thinsp;\u0026plusmn;\u0026thinsp;0.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.99\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e100.0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e96.3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eall (281)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHp\u0026thinsp;+\u0026thinsp;Rp\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e \u003cp\u003e0.86\u0026thinsp;\u0026plusmn;\u0026thinsp;0.03\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.89\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e75.0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e85.2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.20 (60)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e(17)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHp\u0026thinsp;+\u0026thinsp;Rn\u003csup\u003e+\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e \u003cp\u003e0.99\u0026thinsp;\u0026plusmn;\u0026thinsp;0.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.99\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e100.0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e92.6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eall (228)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHp\u0026thinsp;+\u0026thinsp;Rn\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e \u003cp\u003e0.93\u0026thinsp;\u0026plusmn;\u0026thinsp;0.03\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.88\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e87.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e74.1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.20 (60)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e(14)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHn\u003csup\u003e#\u003c/sup\u003e+Rn\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e \u003cp\u003e0.97\u0026thinsp;\u0026plusmn;\u0026thinsp;0.02\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.99\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e100.0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e96.3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eall (68)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHn\u0026thinsp;+\u0026thinsp;Rn\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e \u003cp\u003e0.83\u0026thinsp;\u0026plusmn;\u0026thinsp;0.05\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.92\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e87.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e85.2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.10 (57)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e(15)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHn\u0026thinsp;+\u0026thinsp;Rp\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e \u003cp\u003e0.99\u0026thinsp;\u0026plusmn;\u0026thinsp;0.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.99\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e87.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e100.0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eall (121)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHn\u0026thinsp;+\u0026thinsp;Rp\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e \u003cp\u003e0.89\u0026thinsp;\u0026plusmn;\u0026thinsp;0.04\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.99\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e87.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e100.0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.30 (62)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e(12)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHp\u003csup\u003e*\u003c/sup\u003e+Hn\u003csup\u003e#\u003c/sup\u003e+Rp\u003csup\u003e^\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e \u003cp\u003e0.99\u0026thinsp;\u0026plusmn;\u0026thinsp;0.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.99\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e100.0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e100.0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eall (281)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHp\u0026thinsp;+\u0026thinsp;Hn\u0026thinsp;+\u0026thinsp;Rp\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e \u003cp\u003e0.91\u0026thinsp;\u0026plusmn;\u0026thinsp;0.04\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.97\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e87.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e96.3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.23 (59)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e(21)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHp\u0026thinsp;+\u0026thinsp;Hn\u0026thinsp;+\u0026thinsp;Rn\u003csup\u003e+\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e \u003cp\u003e0.99\u0026thinsp;\u0026plusmn;\u0026thinsp;0.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.99\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e100.0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e88.9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eall (254)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHp\u0026thinsp;+\u0026thinsp;Hn\u0026thinsp;+\u0026thinsp;Rn\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e \u003cp\u003e0.95\u0026thinsp;\u0026plusmn;\u0026thinsp;0.03\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.93\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e87.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e77.8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.20 (58)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e(18)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHn\u0026thinsp;+\u0026thinsp;Rp\u0026thinsp;+\u0026thinsp;Rn\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e \u003cp\u003e0.99\u0026thinsp;\u0026plusmn;\u0026thinsp;0.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.99\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e100.0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e96.3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eall (163)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHn\u0026thinsp;+\u0026thinsp;Rp\u0026thinsp;+\u0026thinsp;Rn\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e \u003cp\u003e0.91\u0026thinsp;\u0026plusmn;\u0026thinsp;0.04\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.99\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e87.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e100.0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.25 (60)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e(17)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHp\u0026thinsp;+\u0026thinsp;Hn\u0026thinsp;+\u0026thinsp;Rp\u0026thinsp;+\u0026thinsp;Rn\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e \u003cp\u003e0.99\u0026thinsp;\u0026plusmn;\u0026thinsp;0.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.99\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e100.0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e100.0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eall (349)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHp\u0026thinsp;+\u0026thinsp;Hn\u0026thinsp;+\u0026thinsp;Rp\u0026thinsp;+\u0026thinsp;Rn\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e \u003cp\u003e0.93\u0026thinsp;\u0026plusmn;\u0026thinsp;0.04\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.99\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e87.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e96.3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.20 (56)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e(26)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"8\"\u003e\u003csup\u003e*\u003c/sup\u003eHp \u0026ndash; HILIC in positive mode\u003c/td\u003e\u003c/tr\u003e \u003ctr\u003e\u003ctd colspan=\"8\"\u003e\u003csup\u003e#\u003c/sup\u003eHn - HILIC in negative mode\u003c/td\u003e\u003c/tr\u003e \u003ctr\u003e\u003ctd colspan=\"8\"\u003e\u003csup\u003e^\u003c/sup\u003eRp \u0026ndash; RPLC in positive mode\u003c/td\u003e\u003c/tr\u003e \u003ctr\u003e\u003ctd colspan=\"8\"\u003e\u003csup\u003e+\u003c/sup\u003eRn - RPLC in negative mode\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e"},{"header":"4. Discussion","content":"\u003cp\u003eA mutation in the NF2 gene was observed in the majority of studied meningiomas, which correlates with literature reporting that this aberration can be detected in over 60% of patients (Lee et al. \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). A wide range of lipids, including long-chain acylcarnitines, phospholipids, sphingolipids, and glycerides, were detected in the experiment presented herein. Previous studies on brain tumors showed a similar range of analytes (Bogusiewicz et al. \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Jarmusch et al. \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Yu et al. \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). What should also be pointed out is that many PC and PE plasmalogens were extracted (Table S2, S3, S4, S5). They may constitute between 20% and 50% of the total phospholipid mass in the brain (Ferreri et al. \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). However, due to high sensitivity to acids, these analytes could be easily disrupted during sample preparation. Moreover, these analytes are antioxidants, so they can potentially be engaged in redox reactions in the sampled tissue, which makes them even more unstable. Therefore, the possibility of extracting these analytes could be related to the phenomenon that enzymatic reactions are quenched by binding analytes to the sorbent, thus reducing the impact of these proteins on lipidome composition (Bogusiewicz et al. \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Consequently, SPME may be a promising sampling method that does not impact plasmalogen stability, similar to the case of oxylipins, which are unstable compounds (Napylov et al. \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eStudies presented herein showed that most phospholipids were downregulated in NF2mt meningiomas. It is surprising due to the known function of merlin in the hippo pathway, which is dysregulated in cancerous cells (Xu et al. \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). The hippo pathway regulates the proliferation and growth of cells, and merlin serves as an inhibitor of this pathway, inhibiting proliferation and growth. Therefore, its lack is expected to be related to cancer growth, which, in turn, is usually related to higher lipid content (Bogusiewicz et al. \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). However, the increase may not have been observed due to the benign character of the studied tumors. The explanation is not clear, so this observation should be a starting point for future studies of this phenomenon. An additional observation is that plasmalogens such as PE P-38:6, PE P-40:6, PC P-34:1, and PC P-36:4 were upregulated in mutant samples and had a high impact on meningioma differentiation (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e,\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e, Table S6). As mentioned earlier, these analytes participate in oxygen reactivates, mitigating damages caused by free radicals (Ferreri et al. \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Antioxidant activity could be crucial in cancer development, making these analytes one of the first potential indicators of changes (Messias et al. \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). Lysophospholipids, which, among others, take part in signaling processes and cell membrane fluidity through the Lands cycle, were not significantly altered in tumors studied tumors (Hishikawa et al. \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2008\u003c/span\u003e). However, chemometric models selected them as discriminatory lipids (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e,\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). The alterations in this group of lipids were reported before, e.g. a higher content of LPC was observed in gliomas with a higher grade and IDH1/2 wildtype lesions, which both have worse clinical outcomes than low-grade and IDH1/2 mutant tumors, respectively (Bogusiewicz et al. \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). It could also indicate that cancerous changes and the lack of suppression in the hippo pathway disrupt lipid turnover in cell membranes. PS should also be mentioned among phospholipids due to their role in apoptosis (Kaynak et al. \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). This group is not highly altered in studied samples, but some individual species, such as PS 36:1, PS 36:2, and PS 40:7, were changed and selected as discriminatory analytes in chemometric models (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e,\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e, Table S6). PS is mainly localized in the inner leaflet of the cell\u0026rsquo;s membrane in normal cells, but due to apoptosis-related changes, PS is translocated to the outer leaflet signaling phagocytic cells to engulf the apoptotic one (Furuta and Zhou \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). However, in cancer cells, PS, due to the activity of flipases, is localized mainly in the outer part of the membrane, leading to immunosuppression and cancer progression (Wang et al. \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). The role of PS in cancer biology is complex and observed downregulation of PS in NF2mt species can be related to a lower capability for apoptosis.\u003c/p\u003e \u003cp\u003eEven though ceramides were not selected as discriminatory analytes, their derivates, including sphingosines, sphingomyelins, and hexylcaramides, were observed as altered in meningiomas with different NF2 statuses. Sphingolipids metabolism, where ceramide is the common substrate, can be integrated into three pathways: sphingomyelins pathway (SM are created), salvage (sphingosine is produced), and modified ceramide hydrolysis where, for instance, hexoylceramide is produced (Li et al. \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). It was reported that alterations in these metabolic pathways could be related to alterations in cancer cell growth and migration, autophagy, and apoptosis processes in neoplastic lesions such as glioblastoma, breast cancer, and liver cancer (Li et al. \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). In research presented herein significant alterations were observed only in sphingomyelins groups, but what is essential is that sphinganine and hexoylceramide were selected as discriminatory analytes (Table S6). It shows their impact on tumor metabolism. The common trend of sphingomyelin changes in NFmt and NFwt tumors was not evident. However, SM participates in creating lipid rafts, and their level increases corresponding to a more rigid structure (Hirano et al. \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Moreover, these analytes are the major source of ceramides, one of the most bioactive lipids (Go\u0026ntilde;i \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). The study presented herein selected sphinganine and sphingosine as discriminatory lipids (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e,\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). They are the building blocks for other sphingolipids (Farley et al. \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Thus, alterations in their level could correspond to further changes in metabolic pathways, e.g. sphinganine takes part in apoptosis, which can be induced by the cleaving of procaspase-3 (Farley et al. \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Hexyl ceramides could also be counted in sphingolipid species, and they were related to the apoptosis.\u003c/p\u003e \u003cp\u003eLastly, acylcarnitine levels with medium and long acyl chains were altered, although they were not selected as significantly altered or discriminatory species (Table S2). Disruption in this group of analytes was widely reported as important in tumor progression due to its correlation with tumor plasticity and the high demand for energy to grow and proliferate (Melone et al. \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). Reports show that acylcarnitine levels were higher in NF2mt meningiomas than wildtype (Bogusiewicz et al. \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). Interestingly the lack of merlin, protein suppressor, and microtubule stabilizer could lead to a higher possibility of developing new foci of cancer development or tumor transformation (Petrilli and Fern\u0026aacute;ndez-Valle \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). It could explain increased energy consumption and elevated levels of lipids as markers of higher lipid turnover and increased fatty acid oxidation in NF2 mutated cells. Moreover, cancerous cell proliferation is related to higher energy consumption. This demand could be fulfilled by glucose metabolism \u0026ndash; Warburg effect or fatty acid oxidation (Melone et al. \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). Although, it was observed that NF2mt cells are characterized by higher dependence on lipid metabolism (Stepanova et al. \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). Increased fatty oxidation is related to an enhanced carnitine shuttle system, which produces higher acylcarnitine levels.\u003c/p\u003e \u003cp\u003eFinally, the selection of potential biomarkers using a particular mode of instrumental analysis resulted in worse differentiation parameters than the model's build on all detected analytes. However, combining more than one chromatography type and ionization mode increased chemometric modeling parameters. It could be related to introducing different information using different types of instrumental analysis. For instance, HILIC enables the separation of hydrophilic analytes such as phospholipids sphingomyelins, and RPLC, on the other hand, detects hydrophobic analytes such as glycerides and ceramides (Cajka and Fiehn \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2014\u003c/span\u003e, \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). The majority of analytes are detected in positive ion mode. However, analytes such as PI, PA, PS, and fatty acids are more commonly observed in negative ionization mode (Ivanova et al. \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2001\u003c/span\u003e). Moreover, adding more variables to the model allowed the preparation of a wider view of the studied tissue and increased the number of variables, which corrected the model parameters.\u003c/p\u003e"},{"header":"5. Conclusions","content":"\u003cp\u003eAmong lipids altered by NF2 mutation were phospholipids, sphingolipids, and acylcarnitines. Interestingly, the trend of upregulation of PC and PE plasmogenes and acylcarnitines in NF2mt meningiomas was observed. Chemometric analysis revealed that based on lipid profiles determined by both the RPLC and HPLC in positive and negative ionization modes made it possible to differentiate the NF2wt meningioma samples from those with NF2mt with the best predictive abilities with a sensitivity and specificity of 100.0%. The same prediction outcome was obtained using the whole set of lipids determined by the combination Hp\u0026thinsp;+\u0026thinsp;Hn and also only Rn. Two combined discriminant models using a reduced number of 12 (Hp\u0026thinsp;+\u0026thinsp;Rp) or 17 (Hp\u0026thinsp;+\u0026thinsp;Rp\u0026thinsp;+\u0026thinsp;Rn) lipids were also found useful, with a sensitivity of 87.5% and a specificity of 100.0%. If there is a need to limit the chromatographic methods to one, the discriminant model using all 42 lipids determined by Rn had the best sensitivity of 100.0% and a specificity of 92.6%.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cp\u003eAUC \u0026ndash; area under curve\u003c/p\u003e\n\u003cp\u003eCer \u0026ndash; Ceramide\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eChE \u0026ndash; cholesterol ester\u003c/p\u003e\n\u003cp\u003eCV \u0026ndash; coefficient of variation\u003c/p\u003e\n\u003cp\u003eDIVA \u0026ndash; discriminating variable test\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eHexCer \u0026ndash; hexoylcramide\u003c/p\u003e\n\u003cp\u003eHILIC \u0026ndash; Hydrophilic interaction chromatography\u003c/p\u003e\n\u003cp\u003eHn - HILIC in negative mode\u003c/p\u003e\n\u003cp\u003eHp \u0026ndash; HILIC in positive mode\u003c/p\u003e\n\u003cp\u003eKLF4 \u0026ndash; Kruppel-like factor 4\u003c/p\u003e\n\u003cp\u003eLC-MS \u0026ndash; Liquid chromatography coupled with mass spectrometry\u003c/p\u003e\n\u003cp\u003eLPC \u0026ndash; Lysophosphatidylcholine\u003c/p\u003e\n\u003cp\u003eLPE \u0026ndash; Lysophosphatidylethanolamine\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eMAD \u0026ndash; median absolute deviation\u003c/p\u003e\n\u003cp\u003eMCCR \u0026ndash; mean correct classification rate\u003c/p\u003e\n\u003cp\u003eMTBE \u0026ndash; Methyl tert-butyl ether\u003c/p\u003e\n\u003cp\u003eNF2m \u0026ndash; NF2 mutant\u003c/p\u003e\n\u003cp\u003eNF2w \u0026ndash; NF2 wildtype\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eNPLC \u0026ndash; normal phase chromatography\u003c/p\u003e\n\u003cp\u003ePA \u0026ndash; Phosphatidic acid\u003c/p\u003e\n\u003cp\u003ePC \u0026ndash; Phosphatidylcholine\u003c/p\u003e\n\u003cp\u003ePE \u0026ndash; Phosphatidylethanolamine\u003c/p\u003e\n\u003cp\u003ePI \u0026ndash; Phosphatidylinositol\u003c/p\u003e\n\u003cp\u003ePLS-DA \u0026ndash; partial least squares\u003c/p\u003e\n\u003cp\u003ePS \u0026ndash; Phosphatidylserine\u003c/p\u003e\n\u003cp\u003eQC \u0026ndash; quality control\u003c/p\u003e\n\u003cp\u003eRn - RPLC in negative mode\u003c/p\u003e\n\u003cp\u003eROC \u0026ndash; receiver operating characteristic\u003c/p\u003e\n\u003cp\u003eRp \u0026ndash; RPLC in positive mode\u003c/p\u003e\n\u003cp\u003eRPLC \u0026ndash; Reversed-phase liquid chromatography\u003c/p\u003e\n\u003cp\u003eSM \u0026ndash; Sphingomyelin\u003c/p\u003e\n\u003cp\u003eSPB \u0026ndash; Sphingosine\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eSPME \u0026ndash; solid-phase microextraction\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eSR \u0026ndash; selectivity ratio\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTP \u0026ndash; target projection\u003c/p\u003e\n\u003cp\u003eTRAF7 \u0026ndash; tumor necrosis factor receptor-associated factor 7\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWHO \u0026ndash; World Health Organization\u003c/p\u003e\n"},{"header":"Declarations","content":"\u003cp\u003eAcknowledgments\u003c/p\u003e\n\u003cp\u003eThe National Science Centre Poland funded lipidomic profiling within research grant No. 2015/18/M/ST4/00059. The National Science Centre Poland funded the genetic tests within research grant No. 2019/33/N/ST4/00286.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe authors acknowledge Supelco/MilliporeSigma for kindly supplying the SPME probes. The authors would like to acknowledge Paulina Zofia Goryńska, Krzysztof Goryński, and Karol Jaroch for their help with samplings.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;Conflict of interest\u003c/p\u003e\n\u003cp\u003eThe authors declare no conflict of interest.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;Author contributions\u003c/p\u003e\n\u003cp\u003eConceptualization: J.B., B.B. M.H.; Methodology: J.B.; Investigation: J.B., M.G.-T., P.S., K.S., A.M. and A.R.; Resources: M.B., B.B., J.F., and M.H.; Data curation: J.B, I. S.; Data analysis and description: I.S.; Writing, original draft preparation: J.B.; Writing, review and editing: B.B.; Visualization: J.B., I.S.; Supervision: B.B.; Project administration: J.B. and B.B.; Funding acquisition: J.B. and B.B. All authors have read and agreed to the published version of the manuscript.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;Ethical Statements\u003c/p\u003e\n\u003cp\u003eThe study was approved by the Bioethical Committee in Bydgoszcz (KB 628/2015).\u003c/p\u003e\n\u003cp\u003eData Availability\u003c/p\u003e\u003cp\u003eThe experimental data that support the findings of this study are available in the https://repod.icm.edu.pl/dataset.xhtml?persistentId=doi:10.18150/R0ZNBL\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eBogusiewicz, J., Burlikowska, K., Łuczykowski, K., Jaroch, K., Birski, M., Furtak, J., et al. (2021). New chemical biopsy tool for spatially resolved profiling of human brain tissue in vivo. \u003cem\u003eScientific Reports\u003c/em\u003e, \u003cem\u003e11\u003c/em\u003e(1), 19522. https://doi.org/10.1038/s41598-021-98973-y\u003c/li\u003e\n\u003cli\u003eBogusiewicz, J., Furtak, J., Birski, M., Soszyńska, K., Majdańska, A., Ryfa, A., et al. (2025). Acylcarnitine Profiling in Meningiomas with Different NF2 Mutation Statuses. \u003cem\u003eInternational Journal of Molecular Sciences\u003c/em\u003e, \u003cem\u003e26\u003c/em\u003e(4), 1570. https://doi.org/10.3390/ijms26041570\u003c/li\u003e\n\u003cli\u003eBogusiewicz, J., Goryńska, P. Z., Gaca, M., Chmara, K., Goryński, K., Jaroch, K., et al. (2020). On-Site Sampling and Extraction of Brain Tumors for Metabolomics and Lipidomics Analysis. \u003cem\u003eJournal of Visualized Experiments\u003c/em\u003e, \u003cem\u003e2020\u003c/em\u003e(159), 159. https://doi.org/10.3791/61260\u003c/li\u003e\n\u003cli\u003eBogusiewicz, J., Kupcewicz, B., Goryńska, P. Z., Jaroch, K., Goryński, K., Birski, M., et al. (2022). Investigating the Potential Use of Chemical Biopsy Devices to Characterize Brain Tumor Lipidomes. \u003cem\u003eInternational Journal of Molecular Sciences\u003c/em\u003e, \u003cem\u003e23\u003c/em\u003e(7), 3518. https://doi.org/10.3390/ijms23073518\u003c/li\u003e\n\u003cli\u003eCajka, T., \u0026amp; Fiehn, O. (2014). Comprehensive analysis of lipids in biological systems by liquid chromatography-mass spectrometry. \u003cem\u003eTrAC Trends in Analytical Chemistry\u003c/em\u003e, \u003cem\u003e61\u003c/em\u003e, 192\u0026ndash;206. https://doi.org/10.1016/j.trac.2014.04.017\u003c/li\u003e\n\u003cli\u003eCajka, T., \u0026amp; Fiehn, O. (2016). Toward Merging Untargeted and Targeted Methods in Mass Spectrometry-Based Metabolomics and Lipidomics. \u003cem\u003eAnalytical Chemistry\u003c/em\u003e, \u003cem\u003e88\u003c/em\u003e(1), 524\u0026ndash;545. https://doi.org/10.1021/acs.analchem.5b04491\u003c/li\u003e\n\u003cli\u003eCudjoe, E., Bojko, B., Delannoy, I., Saldivia, V., \u0026amp; Pawliszyn, J. (2013). Solid-phase microextraction: A complementary InVivo sampling method to microdialysis. \u003cem\u003eAngewandte Chemie - International Edition\u003c/em\u003e, \u003cem\u003e52\u003c/em\u003e(46), 12124\u0026ndash;12126. https://doi.org/10.1002/anie.201304538\u003c/li\u003e\n\u003cli\u003eFarley, S., Stein, F., Haberkant, P., Tafesse, F. G., \u0026amp; Schultz, C. (2024). Trifunctional Sphinganine: A New Tool to Dissect Sphingolipid Function. \u003cem\u003eACS Chemical Biology\u003c/em\u003e, \u003cem\u003e19\u003c/em\u003e(2). https://doi.org/10.1021/acschembio.3c00554\u003c/li\u003e\n\u003cli\u003eFerreri, C., Ferocino, A., Batani, G., Chatgilialoglu, C., Randi, V., Riontino, M. V., et al. (2023). Plasmalogens: Free Radical Reactivity and Identification of Trans Isomers Relevant to Biological Membranes. \u003cem\u003eBiomolecules\u003c/em\u003e, \u003cem\u003e13\u003c/em\u003e(5). https://doi.org/10.3390/biom13050730\u003c/li\u003e\n\u003cli\u003eFuruta, Y., \u0026amp; Zhou, Z. (2023). How do necrotic cells expose phosphatidylserine to attract their predators\u0026mdash;What\u0026rsquo;s unique and what\u0026rsquo;s in common with apoptotic cells. \u003cem\u003eFrontiers in Cell and Developmental Biology\u003c/em\u003e. https://doi.org/10.3389/fcell.2023.1170551\u003c/li\u003e\n\u003cli\u003eGaca-Tabaszewska, M., Bogusiewicz, J., \u0026amp; Bojko, B. (2022). Metabolomic and Lipidomic Profiling of Gliomas\u0026mdash;A New Direction in Personalized Therapies. \u003cem\u003eCancers\u003c/em\u003e, \u003cem\u003e14\u003c/em\u003e(20), 5041. https://doi.org/10.3390/cancers14205041\u003c/li\u003e\n\u003cli\u003eGo\u0026ntilde;i, F. M. (2022). Sphingomyelin: What is it good for? \u003cem\u003eBiochemical and Biophysical Research Communications\u003c/em\u003e. https://doi.org/10.1016/j.bbrc.2022.08.074\u003c/li\u003e\n\u003cli\u003eGoryńska, P. Z., Chmara, K., Kupcewicz, B., Goryński, K., Jaroch, K., Paczkowski, D., et al. (2022). Metabolomic Phenotyping of Gliomas: What Can We Get with Simplified Protocol for Intact Tissue Analysis? \u003cem\u003eCancers\u003c/em\u003e, \u003cem\u003e14\u003c/em\u003e(2), 312. https://doi.org/10.3390/cancers14020312\u003c/li\u003e\n\u003cli\u003eGupta, S., Bi, W. L., \u0026amp; Dunn, I. F. (2018). Medical management of meningioma in the era of precision medicine. \u003cem\u003eNeurosurgical Focus\u003c/em\u003e, \u003cem\u003e44\u003c/em\u003e(4), E3. https://doi.org/10.3171/2018.1.FOCUS17754\u003c/li\u003e\n\u003cli\u003eHirano, K., Kinoshita, M., \u0026amp; Matsumori, N. (2022). Impact of sphingomyelin acyl chain heterogeneity upon properties of raft-like membranes. \u003cem\u003eBiochimica et Biophysica Acta - Biomembranes\u003c/em\u003e, \u003cem\u003e1864\u003c/em\u003e(12). https://doi.org/10.1016/j.bbamem.2022.184036\u003c/li\u003e\n\u003cli\u003eHishikawa, D., Shindou, H., Kobayashi, S., Nakanishi, H., Taguchi, R., \u0026amp; Shimizu, T. (2008). Discovery of a lysophospholipid acyltransferase family essential for membrane asymmetry and diversity. \u003cem\u003eProceedings of the National Academy of Sciences\u003c/em\u003e, \u003cem\u003e105\u003c/em\u003e(8), 2830\u0026ndash;2835. https://doi.org/10.1073/pnas.0712245105\u003c/li\u003e\n\u003cli\u003eIvanova, P. T., Cerda, B. A., Horn, D. M., Cohen, J. S., McLafferty, F. W., \u0026amp; Brown, H. A. (2001). Electrospray ionization mass spectrometry analysis of changes in phospholipids in RBL-2H3 mastocytoma cells during degranulation. \u003cem\u003eProceedings of the National Academy of Sciences of the United States of America\u003c/em\u003e, \u003cem\u003e98\u003c/em\u003e(13). https://doi.org/10.1073/pnas.131195098\u003c/li\u003e\n\u003cli\u003eJarmusch, A. K., Pirro, V., Baird, Z., Hattab, E. M., Cohen-Gadol, A. A., \u0026amp; Cooks, R. G. (2016). Lipid and metabolite profiles of human brain tumors by desorption electrospray ionization-MS. \u003cem\u003eProceedings of the National Academy of Sciences of the United States of America\u003c/em\u003e, \u003cem\u003e113\u003c/em\u003e(6), 1486\u0026ndash;1491. https://doi.org/10.1073/pnas.1523306113\u003c/li\u003e\n\u003cli\u003eKaynak, A., Davis, H. W., Kogan, A. B., Lee, J. H., Narmoneva, D. A., \u0026amp; Qi, X. (2022). Phosphatidylserine: The Unique Dual-Role Biomarker for Cancer Imaging and Therapy. \u003cem\u003eCancers\u003c/em\u003e. https://doi.org/10.3390/cancers14102536\u003c/li\u003e\n\u003cli\u003eKvalheim, O. M. (2010). Interpretation of partial least squares regression models by means of target projection and selectivity ratio plots. \u003cem\u003eJournal of Chemometrics\u003c/em\u003e, \u003cem\u003e24\u003c/em\u003e(7\u0026ndash;8). https://doi.org/10.1002/cem.1289\u003c/li\u003e\n\u003cli\u003eLee, S., Karas, P. J., Hadley, C. C., Bayley V, J. C., Khan, A. B., Jalali, A., et al. (2019). The Role of Merlin/NF2 Loss in Meningioma Biology. \u003cem\u003eCancers\u003c/em\u003e, \u003cem\u003e11\u003c/em\u003e(11), 1633. https://doi.org/10.3390/cancers11111633\u003c/li\u003e\n\u003cli\u003eLendor, S., Hassani, S. A., Boyaci, E., Singh, V., Womelsdorf, T., \u0026amp; Pawliszyn, J. (2019). Solid Phase Microextraction-Based Miniaturized Probe and Protocol for Extraction of Neurotransmitters from Brains in Vivo. \u003cem\u003eAnalytical Chemistry\u003c/em\u003e, \u003cem\u003e91\u003c/em\u003e(7), 4896\u0026ndash;4905. https://doi.org/10.1021/acs.analchem.9b00995\u003c/li\u003e\n\u003cli\u003eLi, R. Z., Wang, X. R., Wang, J., Xie, C., Wang, X. X., Pan, H. D., et al. (2022). The key role of sphingolipid metabolism in cancer: New therapeutic targets, diagnostic and prognostic values, and anti-tumor immunotherapy resistance. \u003cem\u003eFrontiers in Oncology\u003c/em\u003e. https://doi.org/10.3389/fonc.2022.941643\u003c/li\u003e\n\u003cli\u003eLouis, D. N., Perry, A., Wesseling, P., Brat, D. J., Cree, I. A., Figarella-Branger, D., et al. (2021). The 2021 WHO Classification of Tumors of the Central Nervous System: a summary. \u003cem\u003eNeuro-Oncology\u003c/em\u003e, \u003cem\u003e23\u003c/em\u003e(8), 1231\u0026ndash;1251. https://doi.org/10.1093/neuonc/noab106\u003c/li\u003e\n\u003cli\u003eMelone, M. A. B., Valentino, A., Margarucci, S., Galderisi, U., Giordano, A., \u0026amp; Peluso, G. (2018). The carnitine system and cancer metabolic plasticity review-article. \u003cem\u003eCell Death and Disease\u003c/em\u003e, \u003cem\u003e9\u003c/em\u003e, 228. https://doi.org/10.1038/s41419-018-0313-7\u003c/li\u003e\n\u003cli\u003eMessias, M. C. F., Mecatti, G. C., Priolli, D. G., \u0026amp; De Oliveira Carvalho, P. (2018). Plasmalogen lipids: Functional mechanism and their involvement in gastrointestinal cancer. \u003cem\u003eLipids in Health and Disease\u003c/em\u003e. https://doi.org/10.1186/s12944-018-0685-9\u003c/li\u003e\n\u003cli\u003eNapylov, A., Reyes-Garces, N., Gomez-Rios, G., Olkowicz, M., Lendor, S., Monnin, C., et al. (2020). In Vivo Solid-Phase Microextraction for Sampling of Oxylipins in Brain of Awake, Moving Rats. \u003cem\u003eAngewandte Chemie - International Edition\u003c/em\u003e, \u003cem\u003e59\u003c/em\u003e(6), 2392\u0026ndash;2398. https://doi.org/10.1002/anie.201909430\u003c/li\u003e\n\u003cli\u003eNowosielski, M., Galldiks, N., Iglseder, S., Kickingereder, P., Von Deimling, A., Bendszus, M., et al. (2017). Diagnostic challenges in meningioma. \u003cem\u003eNeuro-Oncology\u003c/em\u003e, \u003cem\u003e19\u003c/em\u003e(12). https://doi.org/10.1093/neuonc/nox101\u003c/li\u003e\n\u003cli\u003ePan, M., Qin, C., \u0026amp; Han, X. (2021). Lipid Metabolism and Lipidomics Applications in Cancer Research. In \u003cem\u003eAdvances in Experimental Medicine and Biology\u003c/em\u003e (Vol. 1316, pp. 1\u0026ndash;24). https://doi.org/10.1007/978-981-33-6785-2_1\u003c/li\u003e\n\u003cli\u003ePetrilli, A. M., \u0026amp; Fern\u0026aacute;ndez-Valle, C. (2016). Role of Merlin/NF2 inactivation in tumor biology. \u003cem\u003eOncogene\u003c/em\u003e, \u003cem\u003e35\u003c/em\u003e(5), 537\u0026ndash;548. https://doi.org/10.1038/onc.2015.125\u003c/li\u003e\n\u003cli\u003eRajalahti, T., Arneberg, R., Kroksveen, A. C., Berle, M., Myhr, K. M., \u0026amp; Kvalheim, O. M. (2009). Discriminating variable test and selectivity ratio plot: Quantitative tools for interpretation and variable (biomarker) selection in complex spectral or chromatographic profiles. \u003cem\u003eAnalytical Chemistry\u003c/em\u003e, \u003cem\u003e81\u003c/em\u003e(7). https://doi.org/10.1021/ac802514y\u003c/li\u003e\n\u003cli\u003eStepanova, D. S., Semenova, G., Kuo, Y. M., Andrews, A. J., Ammoun, S., Hanemann, C. O., \u0026amp; Chernoff, J. (2017). An essential role for the tumor-suppressor merlin in regulating fatty acid synthesis. \u003cem\u003eCancer Research\u003c/em\u003e, \u003cem\u003e77\u003c/em\u003e(18). https://doi.org/10.1158/0008-5472.CAN-16-2834\u003c/li\u003e\n\u003cli\u003eWang, W., Wu, S., Cen, Z., Zhang, Y., Chen, Y., Huang, Y., et al. (2022). Mobilizing phospholipids on tumor plasma membrane implicates phosphatidylserine externalization blockade for cancer immunotherapy. \u003cem\u003eCell Reports\u003c/em\u003e, \u003cem\u003e41\u003c/em\u003e(5). https://doi.org/10.1016/j.celrep.2022.111582\u003c/li\u003e\n\u003cli\u003eXu, D., Yin, S., \u0026amp; Shu, Y. (2024). NF2: An underestimated player in cancer metabolic reprogramming and tumor immunity. \u003cem\u003enpj Precision Oncology\u003c/em\u003e, \u003cem\u003e8\u003c/em\u003e(1), 133. https://doi.org/10.1038/s41698-024-00627-5\u003c/li\u003e\n\u003cli\u003eYu, D., Xuan, Q., Zhang, C., Hu, C., Li, Y., Zhao, X., et al. (2020). Metabolic Alterations Related to Glioma Grading Based on Metabolomics and Lipidomics Analyses. \u003cem\u003eMetabolites\u003c/em\u003e, \u003cem\u003e10\u003c/em\u003e(12), 478. https://doi.org/10.3390/metabo10120478\u003c/li\u003e\n\u003cli\u003eZąbek, A., Stanimirova, I., Deja, S., Barg, W., Kowal, A., Korzeniewska, A., et al. (2015). Fusion of the 1H NMR data of serum, urine and exhaled breath condensate in order to discriminate chronic obstructive pulmonary disease and obstructive sleep apnea syndrome. \u003cem\u003eMetabolomics\u003c/em\u003e, \u003cem\u003e11\u003c/em\u003e(6). https://doi.org/10.1007/s11306-015-0808-5\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"metabolomics","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"mebo","sideBox":"Learn more about [Metabolomics](http://link.springer.com/journal/11306)","snPcode":"11306","submissionUrl":"https://submission.nature.com/new-submission/11306/3","title":"Metabolomics","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"meningioma, brain tumor, SPME, lipidomics, NF2, merlin","lastPublishedDoi":"10.21203/rs.3.rs-6490941/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6490941/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eIntroduction: Meningiomas are mainly benign brain tumors, but they can evolve to higher grades. The phenomena of these changes are not well-known. Therefore, more basic research is needed. This study attempted to assess the lipidome profile in meningiomas harboring different NF2 mutation statuses (wildtype and mutated). Solid-phase microextraction (SPME) probes were used to sample and extract the metabolites and reduce the invasiveness of lipidomic analysis.\u003c/p\u003e\n\u003cp\u003eObjectives: This study aimed to select the set of lipids distinguishing meningiomas with different genotypes using two chromatography methods (hydrophilic interaction chromatography (HILIC) and reversed-phase chromatography (RPLC) in two ionization modes.\u003c/p\u003e\n\u003cp\u003eMethods: Brain tumors were obtained during neurosurgical procedures. Then, sampling using SPME fibers was performed directly after the lesion excision. After collecting the whole batch of samples, desorption using an isopropanol-methanol solution was performed. Subsequently, instrumental analysis was carried out using liquid chromatography coupled with high-resolution mass spectrometry. The remaining part of the lesion was stored as paraffin tissue blocks, and then genetic testing was performed to determine the presence of mutations in the NF2 gene.\u003c/p\u003e\n\u003cp\u003eResults: Genetic profiling of meningiomas revealed that most lesions had a mutation in the NF2 gene. A wide range of analytes was extracted from studied tumors using SPME probes, but it was possible to select a set of 26 lipids crucial in tumor differentiation. It was also observed that a combination of analytes detected in more than one analysis mode increased the differentiation of mutant and wildtype samples, which was presented by the high sensitivity and specificity of the prepared models.\u003c/p\u003e\n\u003cp\u003eConclusions: SPME coupled liquid chromatography and mass spectrometry, can be successfully applied to the screening of lipids in meningiomas with different NF mutation statuses.\u003c/p\u003e","manuscriptTitle":"Lipidomic profile of meningiomas harboring different NF2 mutation status","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-04-29 18:21:53","doi":"10.21203/rs.3.rs-6490941/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-08-07T01:43:39+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-08-03T19:02:23+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-07-10T15:33:03+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"51763740142462075458941424351337773186","date":"2025-06-12T20:53:58+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"18056430768973570550030783657828506742","date":"2025-05-01T18:07:20+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-04-28T19:03:10+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-04-21T00:28:52+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-04-21T00:27:24+00:00","index":"","fulltext":""},{"type":"submitted","content":"Metabolomics","date":"2025-04-20T19:57:48+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"metabolomics","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"mebo","sideBox":"Learn more about [Metabolomics](http://link.springer.com/journal/11306)","snPcode":"11306","submissionUrl":"https://submission.nature.com/new-submission/11306/3","title":"Metabolomics","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"095f6529-7c26-4bc6-9111-c176bf5250e5","owner":[],"postedDate":"April 29th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-01-09T20:08:42+00:00","versionOfRecord":[],"versionCreatedAt":"2025-04-29 18:21:53","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6490941","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6490941","identity":"rs-6490941","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}