Citations

Clear

9383 citation(s) found for "Thomas, PS".

International Reference Values for Surgical Outcomes of Total Pancreatectomy
Berchtold, Caroline , Kuemmerli, Christoph, Li, Zhihao , Eden, Janina , Wu, Pengfei , Cecchetto, Sara Nicole, Kendrick, Michael L. , Pecorelli, Nicolò , Guarneri, Giovanni , Farooqui, Waqas , Tschuor, Christoph , Burgdorf, Stefan Kobbelgaard , Mühlhäusser, Julia , Gass, Jörn‐Markus , Koh, Ye-Xin, Seip, Tomas, Santibañes, Martín de , Todeschini, Letizia , Marchegiani, Giovanni, Belfil, Nadiya , Boggi, Ugo, Cappelle, Marie , Koerkamp, Bas Groot , Dutkowski, Philipp , Hackert, Thilo , Truty, Mark J. , Group, Outcomes for Total Pancreatectomy, Valle, Valentina , Giulianotti, Pier Cristoforo , Pessaux, Patrick, Thapa, Prabin , Kuzman, Matta , Pandanaboyana, Sanjay, He, Jin , Kendrick, Michael, Truty, Mark, Li, Zhihao , Groot Koerkamp, Bas , Müller, Philip C , Breuer, Eva , Vallorani, Alessia , Hansen, Carsten Palnæs , Guidetti, Cristiano , Campbell, Brady A. , Cecchetto, Sara Nicole, Gudmundsdottir, Hallbera , Starlinger, Patrick , Goh, Brian K P , Koh, Ye-Xin , Rebelo, Artur , Kleeff, Jorg H. , Marchegiani, Giovanni , Lesurtel, Mickaël , Machado, Marcel Autran C. , Kauffmann, Emanuele F. , Benedetto, Fabrizio Di , Roberts, Keith John , Nevler, Avinoam , Lavu, Harish , Nickel, Felix , He, Jing , Falconi, Massimo , Billeter, Adrian T. , Müller-Stich, Beat Peter , Group, Outcomes for Total Pancreatectomy, Halle-Smith, James Michael , Giannone, Fabio , Pessaux, Patrick , Sánchez‐Velázquez, Patricia , Maharjan, Dhiresh Kumar , Torres, Orlando Jorge Martins

Mayo Clinic

JAMA Surgery | November 12, 2025 | Volume: 161 | Issue: 1 | Pages: 50

10.1001/jamasurg.2025.4941 | PMID 41222926 Get It

Importance Total pancreatectomy (TP) is indicated for advanced pancreatic cancer or multifocal tumors. Furthermore, TP may be performed to avoid the risk of pancreatic fistula in selected patients to improve the perioperative risk profile. Objective To define reference values for TP based on a low-risk cohort treated at expert centers. Design, Setting, and Participants This multicenter study analyzed outcomes from patients undergoing primary TP for malignant or benign lesions from 25 international expert centers from January 2017 to November 2023. Low-risk patients undergoing TP (LR-TP) were without vascular resections or significant comorbidities. Exposures TP. Main Outcomes and Measures Twenty reference values were derived from the 75th or the 25th percentile of the median values of all centers. Outcomes of LR-TP were compared with a cohort of TP with vascular resection, TP due to high-risk pancreatic anastomosis, and the benchmark values for low-risk pancreatoduodenectomy. Results Of 994 patients, 333 (33.5%; median [IQR] age, 66 [58-72] years; 171 male [51.4%]) qualified as the LR-TP cohort. Reference values included blood loss (≤1000 mL), major complications (≤37%), 3-month postoperative mortality (<6%), and retrieved lymph nodes (≥29). Compared with TP with vascular resections, reference cutoffs were not met for major complications (51% vs LR-TP ≤37%) and 90-day mortality (11% vs LR-TP ≤6%). For TP due to high-risk anastomosis, failure to rescue rate (38% vs ≤6%) and 90-day mortality (11% vs LR-TP ≤6%) were not met. Compared with pancreatoduodenectomy, reference values for postoperative mortality were 3 times higher for LR-TP (≤2% vs ≤6%) and less for resected lymph nodes (≥16 vs ≥29). Conclusions and Relevance This case-control study provided global reference values for TP, indicating significantly higher postoperative morbidity and mortality compared with pancreatoduodenectomy. Perioperative morbidity of TP was especially increased in patients with vascular resections. These reference values can serve for quality control of pancreatic surgery.
Integrated analysis of the safety of fecal microbiota, live-jslm in adults with recurrent Clostridioides difficile infection from five prospective clinical trials: an update
Lee, Christine , Feuerstadt, Paul , Louie, Thomas , Bancke, Lindy, Guthmueller, Beth, Harvey, Adam Charles , Hoeyer, Frederikke, Orenstein, Robert , Khanna, Sahil , Feuerstadt, Paul , Harvey, Adam, Dubberke, Erik R.

Mayo Clinic

Therapeutic Advances in Gastroenterology | November 12, 2025 | Volume: 18 | Issue: No issue available | Pages: No page available

10.1177/17562848251395566 | PMID 41245385 Get It

Background: Fecal microbiota, live-jslm (RBL) is approved in the United States and Canada for prevention of recurrent Clostridioides difficile infection (rCDI) in adults following standard-of-care (SOC) antibiotic treatment. Objectives: Provide an updated integrated safety analysis, incorporating final safety data from Punch CD3-OLS. Design: Safety data were combined from five RBL trials: three phase II and two phase III trials. Methods: Adult participants had documented rCDI and completed SOC therapy before receiving one or two doses of RBL or placebo, rectally administered as one treatment course. Treatment-emergent adverse events (TEAEs) were recorded for ⩽6 months. Results: TEAEs were reported in 70.9% (845/1192) of RBL recipients; most TEAEs were mild to moderate and gastrointestinal in nature. Most serious TEAEs were related to preexisting conditions or CDI. There was no clustering of serious TEAEs. Most TEAEs leading to death were related to preexisting conditions. Conclusion: Overall, data demonstrate RBL has a favorable 6-month safety profile. Trial registration: ClinicalTrials.gov: NCT01925417; NCT02299570; NCT02589847; NCT03244644; NCT03931941.
394 Global Cytopathology in Practice: Medical Student Experience with Cervical Pap Screening in Cusco, Peru
Schroeder, MC, Escalante, Erika, Ferris, Daron, Schroeder, Mariah, Reynolds, Jordan , Thompson, Thomas

Mayo Clinic

American Journal of Clinical Pathology | November 12, 2025 | Volume: 164 | Issue: Supplement_1 | Pages: No page available

10.1093/ajcp/aqaf121.079 Get It

Abstract Introduction/Objective Cervical cancer remains a global health concern, with significantly higher incidence in countries like Peru compared to the U.S. Cytopathology plays a vital role in early detection and prevention, making it an important area of focus for aspiring pathologists. Methods/Case Report A medical student with an interest in pathology participated in a global health rotation in Cusco, Peru, at CerviCusco, a nonprofit clinic dedicated to cervical cancer care. Under the guidance of a trained cytotechnologist, the student engaged in hands-on cytopathology work, including processing and screening SurePath (Franklin Lakes, NJ) slides, and reviewing abnormal cases. The student had access to a microscope and dedicated time developing microscopy skills and learning cervical cytomorphology, with emphasis on identifying significant abnormalities. Results NA Conclusion This experience demonstrates the educational value of incorporating global health electives into medical training, especially in resource-limited settings with high disease burden. Exposure to such environments enhances foundational diagnostic skills and fosters an early appreciation for global health disparities. Expanding similar opportunities can enrich pathology education and inspire future contributions to international health efforts.
Meniscal Repair in the Setting of Revision Anterior Cruciate Ligament Reconstruction: 6-Year Follow-up Results From the MARS Cohort
Group;, MARS, Fox, Jake A. , Huston, Laura J. , Haas, Amanda K. , Cooper, Daniel E. , Dunn, Warren R. , Lantz, Brett A., Amendola, Annunziato , Annunziata, Christopher C., Arciero, Robert A. , Bach, Bernard R. , Baker, Champ L., Bartolozzi, Arthur R. , Baumgarten, Keith M. , Berg, Jeffrey H., Bernas, Geoffrey A., Brockmeier, Stephen F. , Bush‐Joseph, Charles A., Butler, J. Brad, Carey, James L. , Carpenter, James E. , Cole, Brian J. , Cooper, Jonathan M. , Creighton, R. Alexander , Flanigan, David C. , Frederick, Robert W., Ganley, Theodore J. , Gatt, Charles J. , Gecha, Steven R., Hame, Sharon L. , Hannafin, Jo A., Harner, Christopher D. , Harris, Norman Lindsay, Hechtman, Keith S., Hershman, Elliott B., Hoellrich, Rudolf G., Johnson, David C. , Kaeding, Christopher C., Kamath, Ganesh V. , Klootwyk, Thomas E., Levy, Bruce A. , Benjamin, C. , Maiers, G. Peter, Marx, Robert G. , Matava, Matthew J. , Mathien, Gregory M., McCormack, Robert G. , Miller, Bruce S., Nissen, Carl W. , O’Neill, Daniel F., Parker, Richard D., Purnell, Mark L., Ramappa, Arun J. , Rauh, Michael A. , Rettig, Arthur C., Sekiya, Jon K., Shea, Kevin G. , Slauterbeck, James R., Smith, Matthew V. , Svoboda, Col. Steven J., Taft, Timothy N., Tenuta, Joachim J., Tingstad, Edwin M., Vidal, Armando F. , Viskontas, Darius G., White, Richard A., Williams, James S., Wolcott, Michelle L., York, James J. , Wright, Rick W. , Svoboda, Steven J. , Group;, MARS, Pennings, Jacquelyn S. , Allen, Christina R. , DeBerardino, Thomas M. , Spindler, Kurt P , Stuart, Michael J., Brophy, Robert H. , Carpenter, James E., Cox, Charles L. , David, Tal S. , Flanigan, David C. , Giffin, J. Robert , Johnson, David C., Johnson, Timothy S. , Jones, Morgan H. , Kaeding, Christopher C. , Marx, Robert G. , McAllister, David R. , McCarty, Eric C. , Miller, Bruce S., Owens, Brett D. , Sherman, Orrin H. , Spang, Jeffrey T. , Vidal, Armando F. , Wolf, Brian R.

Mayo Clinic

The American Journal of Sports Medicine | November 12, 2025 | Volume: 53 | Issue: 14 | Pages: 3435-3445

10.1177/03635465251387333 | PMID 41220248 Get It

Background: Meniscal preservation has been demonstrated to contribute to long-term knee health and has been a successful intervention in isolation and in patients with anterior cruciate ligament reconstruction (ACLR). The long-term results of meniscal repair in the setting of revision ACLR have yet to be documented. Purpose: To report the incidence of meniscal repair failures at the 6-year follow-up in a cohort of patients who underwent concurrent revision ACLR and primary meniscal repair. Study Design: Prospective cohort study; Level of evidence, 2. Methods: All revision ACLRs with concomitant primary meniscal repair cases from a multicenter group between 2006 and 2011 were selected. Six-year follow-up was obtained to determine whether any subsequent surgery had occurred since their initial revision ACLR. If so, operative reports were obtained, whenever possible, to verify pathological condition and treatment. Results: In total, 221 patients from 1234 revision ACLRs underwent concurrent primary meniscal repairs (18% of the cohort). There were 238 repairs performed: 173 medial and 65 lateral. The majority of these repairs (n = 181; 76%) were performed with an all-inside technique. Six-year surgical follow-up was obtained in 77% (171/221) of the cohort, or 189 of 238 (79%) of the repairs (136 medial, 53 lateral). The meniscal repair failure rate, defined as reoperation, was 16% (31/189) at 6 years. Of the 31 failures, 28 were medial (24 all-inside, 4 inside-out; 28/136 = 20.6% failure rate) and 3 were lateral (2 all-inside, 1 inside-out; 3/53 = 5.7% failure rate). Three medial failures were treated in conjunction with a subsequent repeat revision ACLR. Medial tears underwent reoperation for failure at a significantly higher rate than lateral tears (20.6% vs 5.7%; P = .01) and had a significantly shorter survival time compared with lateral tears ( P = .02). No difference was found between the failure and nonfailure groups when it came to tear type, tear length, repair technique utilized, suture/implant type, or number of sutures used between the 2 groups. Conclusion: Meniscal repair in the revision ACLR setting has a 16% failure rate at 6 years. Failure rates for medial tears (20.6%) were found to be higher than that for lateral tears (5.7%), which aligns with previous studies in both the revision and primary ACLR setting.
Transcriptional pause release at enhancers mediates cell identity
Ekstrom, Thomas L. , Johnsen, Steven A. , Johnsen, Steven A

Mayo Clinic

NAR Cancer | November 11, 2025 | Volume: 7 | Issue: 4 | Pages: No page available

10.1093/narcan/zcaf039 | PMID 41230003 Get It

Abstract This short review highlights recent findings related to transcription factor-mediated RNA polymerase II (Pol II) pause release at enhancers and its role in regulating cell identity. We discuss how the transcriptional state of Pol II, either paused or released, shapes enhancer–promoter interactions, thereby influencing gene expression programs that define cell identity. We further summarize current knowledge of DNA-binding transcription factors that regulate Pol II pause release at enhancers and explore how this process influences cell identity. We propose a novel concept of therapy-induced molecular amnesia, where temporary inhibition of enhancer-proximal Pol II pause release and enhancer RNA transcription affects enhancer–promoter interactions differently in distinct cellular contexts. In some contexts, preservation of enhancer–promoter interactions allows rapid transcriptional reactivation and maintenance of cell identity upon recovery. In other contexts, repeated inhibition of enhancer RNA transcription leads to the permanent dissociation of enhancer–promoter interactions and loss of cell identity. Our model suggests that manipulating Pol II pause release at enhancers could selectively reverse certain aggressive tumor cell identities, limit tumor cell plasticity, and improve therapy responsiveness.
TMIC-83. Targeting Lipid Metabolism in GBM: A Systems Biology Strategy to Overcome Treatment-Resistant Persister Cells and Their Myeloid Cell-Mediated Support
Chakraborty, Avirup , Yang, Changlin , Feier, Diana , Silver, Aryeh , Searl, Nyla, Anica, Miruna, Pittu, Avinash, West, Illeana, Sobanjo, Olusegun O., Hodge, Ethan, Engelbart, Mia, Sengupta, Srabasti, Thomas, Nagheme, Rahman, Maryam, Mitchell, Duane A. , Sarkisian, Matthew R. , Chakraborty, Avirup , Feier, Diana, Sengupta, Srabasti, Roemeling, Christina A. Von , Huang, Jiangping, Harrison, Jeffrey K. , Deleyrolle, Loic P.

Mayo Clinic

Neuro-Oncology | November 11, 2025 | Volume: 27 | Issue: Supplement_5 | Pages: v464-v464

10.1093/neuonc/noaf201.1838 Get It

Abstract INTRODUCTION Glioblastoma (GBM) contains metabolically distinct cell lineages: treatment-sensitive cells (TSCs) dependent on glycolysis and treatment-resistant persister cells (TRPCs) reliant on lipid metabolism. Our integrative systems biology approach, combining spatial multi-omics and targeted analyses, uncovers lipid acquisition in TRPCs as a pivotal driver of GBM progression which is supported by a retrospective study of statin-treated patients, providing clinical evidence for the therapeutic potential of targeting lipid metabolism in GBM patients stratified based on high TRPC signatures. METHODS We applied spatial transcriptomics and proteomics combined with 3D reconstruction of the tumor microenvironment to map TRPC- and TSC-enriched regions. Immunohistochemically stained tissues were analyzed using CellProfiler and our custom SNAQ algorithm for single-cell neighborhood analysis. Lipid transfer from Tumor-Associated Myeloid Cells (TAMCs) to TRPCs was tracked via flow cytometry, holotomography, and time-lapse imaging. Genetic and pharmacological targeting of CCR2, CSF1R, and FABP3 was performed in vitro and in vivo. Finally, a retrospective analysis of 4,085 GBM patients assessed the impact of statin treatment in relation to TRPC signatures. RESULTS TRPCs exploit recruited TAMCs to acquire lipids, supporting their survival and immune evasion. These interactions are mediated by CCL2/7–CCR2 and CSF1–CSF1R signaling. Disruption of TRPC-TAM lipid homeostasis significantly remodeled the immune microenvironment, reduced tumor growth, and extended survival in preclinical models. Patients with high TRPC signatures typically had poor outcomes; however, retrospective analysis revealed that statin treatment was associated with improved survival in this subgroup. These findings position lipid trafficking as a metabolic vulnerability in GBM and highlight the therapeutic relevance of targeting TRPC–TAMC interactions. CONCLUSION Our study reveals lipid trafficking to TRPCs as a key mechanism driving GBM progression. Targeting TAMC-mediated lipid transfer disrupts immune suppression and tumor growth. Transcriptomic profiling may guide patient stratification, identifying those likely to benefit from lipid metabolism-targeted therapies such as statins.
IMG-87. Differentiation of IDH-Wildtype Glioblastoma and Primary Central Nervous System Lymphoma Using 3D Deep Learning on MRI
Moassefi, Mana, Decker, Paul A. , Kosel, Matthew, Ruff, Michael W. , Farooq, Umar , Jenkins, Robert B. , Eckel‐Passow, Jeanette E. , Moassefi, Mana , Conte, Gian Marco , Decker, Paul A. , Burns, Terry C. , Farooq, Umar, Derdeyn, Colin P. , Habermann, Thomas M. , Cerhan, James R. , Slager, Susan L. , Vaubel, Rachael A. , Lachance, Daniel Honore , Erickson, Bradley J. , Tobin, W. Oliver

Mayo Clinic

Neuro-Oncology | November 11, 2025 | Volume: 27 | Issue: Supplement_5 | Pages: v294-v294

10.1093/neuonc/noaf201.1166 Get It

Abstract Glioblastoma, IDH-wildtype (GBM), and primary central nervous system lymphoma (PCNSL) are two aggressive brain tumors with distinct treatment strategies but overlapping imaging features on MRI. Accurate preoperative differentiation remains a clinical challenge, due to their overlapping imaging features, even when advanced MRI techniques are employed. This study explored deep learning approaches leveraging multi-sequence MRI to distinguish GBM from PCNSL. Models were developed utilizing 146 PCNSL cases, which were matched to146 GBM by age, sex and year of diagnosis. A 3D-DenseNet121 convolutional neural network was trained using two analytical strategies: (i) loss-based model selection and (ii) ensemble prediction using five-fold cross-validation. Model performance was evaluated on an independent validation set of 256 GBM and 37 PCNSL from the same institution as well as 36 PCNSL cases from an external institution. All MRIs were preprocessed using the FeTS pipeline. Using the independent validation cohort, area under the receiver operating characteristic curve (AUC) and 95% confidence intervals (CI) were computed overall and stratified by demographic characteristics. Using both T1 post-contrast (T1Gd) and T2-weighted images, the loss-based model achieved AUC=0.83 (95% CI: 0.77-0.88) and the ensemble model yielded an AUC=0.82 (95% CI: 0.76-0.89). Model performance was similar across males and females; however, model performance was better for subjects older than 50 years of age versus younger than 50. When using only T1Gd images, the loss-based model dropped to AUC=0.55 (95% CI: 0.48-0.63), whereas the ensemble model had AUC=0.79 (95% CI: 0.72-0.85). Models trained with only T2 images achieved AUCs of 0.67 (95% CI: 0.60-0.73) (loss-based) and 0.75 (95% CI: 0.69-0.82) (ensemble). These results highlight the added value of the multi-sequence input. Our results confirm the feasibility of using automated deep learning models to differentiate GBM from PCNSL. Notably, the model generalized well to external MRI data, emphasizing its potential utility as a diagnostic aid in real-world clinical workflows.
PATH-104. Comparison of clinical, radiographic and genomic alterations between classic and molecular glioblastoma, IDH-wildtype (updated report)
Decker, Paul, Kollmeyer, Thomas , Kosel, Matthew , Odukoya, AbdulLateef, Kizilbash, Sani H. , Maqbool, Asad, Ida, Cris, Jenkins, Robert B. , Passow, Jeanette E. Eckel, Kollmeyer, Thomas M. , Kizilbash, Sani H , Lachance, Daniel Honore , Drucker, Kristen , Praska, Corinne , Maqbool, Asad, Shafqat, Shameel , Tobin, W. Oliver , Burns, Terry C. , Parney, Ian F. , Galanis, Evanthia , Giannini, Caterina , Vaubel, Rachael A.

Mayo Clinic

Neuro-Oncology | November 11, 2025 | Volume: 27 | Issue: Supplement_5 | Pages: v266-v266

10.1093/neuonc/noaf201.1056 Get It

Abstract Glioblastoma (GBM) diagnosis integrates morphology and tumor molecular alterations. Patients with classic morphology are designated as histologic GBM (hGBM). Patients without classic morphology harboring TERT promoter mutation (TERTp), or epidermal growth factor receptor amplification (EGFRamp), or combined gain of chromosome 7 and loss of 10 (+7/-10) designate molecular GBM (mGBM). We compared radiographic features, genomic alterations, and clinical outcome between hGBM and mGBM. 109 mGBM Mayo Clinic cases diagnosed between 1993-2021 were matched to 101 cGBM by age, sex, year of diagnosis, and TERT promoter (TERTp) status. Clinical and radiographic data abstraction was blinded. Progression-free and overall survival were compared across groups in the subset of patients who received standard of care. Contrast enhancement was present on index MRI (MRI at diagnosis) for 98% (96/98) hGBM and 50% (54/107) mGBM (p<0.0001) patients. mGBM tumors harbored fewer total copy number alterations (median of 16.5 vs 20, p=0.029) and were less likely to have CDKN2A/B homozygous deletion (56% vs 77%, p=0.005). Among the 87 mGBM and 82 hGBM receiving standard of care, median overall survival was longer for mGBM versus hGBM (2.24 vs 1.62 years, p=0.016). mGBM patients with +7/-10 (HR=2.60, p=0.035) or CDKN2A/B (HR=3.61, p=0.002) had poorer overall survival compared to mGBM patients without the corresponding alteration. On index MRI mGBM patients without contrast enhancement had better overall survival compared to hGBM (HR=0.50; p=0.003) whereas there was no survival difference between mGBM with enhancement and hGBM. mGBM with enhancement on index MRI likely represent undersampled, or early / evolving hGBM. mGBM patients with non-enhancing MRI, younger than 60, with long symptom profile including seizures, and some combination of non-anaplastic morphology, TERTp only, no homozygous deletion of CDKN2A/B, and low copy number complexity may represent other IDHwt glioma entities. Caution is advised in counseling, management, discussing prognosis, and inclusion in clinical trials.
PATH-70. Neuropath-IHC: a deep neural network for virtual immunohistochemistry from digital whole slide images of CNS tumors
Lalchungnunga, Fnu, Hoang, Danh-Tai, Abdullaev, Zied , Li, Bochong , Nasrallah, MacLean P. , Quezado, Martha , Chung, Hye‐Jung, Yefet, Leeor, Zadeh, Gelareh , Aldape, Kenneth, Dampier, Christopher H. , Shulman, Eldad David , Li, Bochong , Luo, Zhirui , Singh, Omkar, Wu, Zhichao , Pearce, Thomas M. , Marker, Daniel F. , Horbinski, Craig , Lucas, Calixto‐Hope G. , Cimino, Patrick J. , Yefet, Leeor S. , Brandner, Sebastian , Ruppin, Eytan

Mayo Clinic

Neuro-Oncology | November 11, 2025 | Volume: 27 | Issue: Supplement_5 | Pages: v257-v257

10.1093/neuonc/noaf201.1022 Get It

Abstract Computer vision now enables deep learning models to link histopathologic features to molecular features in a quantitative manner not previously achievable by human pathologists. However, human interpretability is often limited. We trained a deep neural network to predict gene expression from digital whole slide images (WSIs) using 848 central nervous system (CNS) tumors with paired WSI and RNA-sequencing data. We used inferred RNA expression levels as a surrogate for protein expression of the CNS tumor lineage markers GFAP, OLIG2, and SSTR2, as well as the proliferation marker MKI67 (Ki67). We established thresholds for categorizing the inferred expression levels as positive or negative based on levels observed in cross-validation testing in select tumor types to approximate routine clinical interpretation of immunohistochemical (IHC) staining. We tested the sensitivity and specificity of our ‘virtual’ IHC on an independent, multi-institutional cohort of over 2,000 CNS tumors with objective diagnostic labels derived from DNA methylation-based tumor classification. As a dichotomous variable, inferred GFAP expression showed a sensitivity of 74% and a specificity of 99% as a glial marker in a cohort of gliomas and meningiomas. In the same cohort, OLIG2 showed a sensitivity of 75% and a specificity of 97% as a glial marker, while SSTR2 showed a sensitivity of 83% and a specificity of 97% as a marker of meningioma. In a cohort of ependymomas and gliomas, virtual IHC for OLIG2 showed a sensitivity of 75% and a specificity of 82% in distinguishing ependymomas from gliomas. Finally, in a cohort of gliomas, inferred expression of MKI67 demonstrated a trend consistent with what would be expected by actual IHC, with increasing MKI67 expression with increasing tumor grade. Our model provides the basis for a human interpretable and clinically applicable deep neural network to aid human pathologists in the diagnosis and grading of CNS tumors.
IMG-38. Differentiating adult diffuse IDH-mutant glioma from molecular glioblastoma IDH-wildtype using deep learning on MRI
Moassefi, Mana, Decker, Paul A. , Kosel, Matthew , Kollmeyer, Thomas M. , Guido, Fiorella, Ruff, Michael W. , Jenkins, Robert B. , Eckel‐Passow, Jeanette E. , Kollmeyer, Thomas , Moassefi, Mana , Conte, Gian Marco , Decker, Paul A. , Guido, Fiorella, Burns, Terry C. , Galanis, Evanthia , Vaubel, Rachael A. , Ida, Cristiane M. , Giannini, Caterina , Lachance, Daniel Honore , Oliver, Bradley, Tobin, W. Oliver

Mayo Clinic

Neuro-Oncology | November 11, 2025 | Volume: 27 | Issue: Supplement_5 | Pages: v281-v281

10.1093/neuonc/noaf201.1117 Get It

Abstract The 2021 WHO classification redefined glioblastoma IDH-wildtype (GBM), integrating histopathological and tumor molecular features. Diffuse astrocytic glioma lacking histologic features of necrosis and microvascular proliferation are classified as (molecular) GBM if the tumor is IDH- and H3-wildtype and has one or more additional tumor molecular alterations: TERT promoter mutation, EGFR amplification, or +7/-10. Due to the absence of histopathological features of classic GBM and presentation without enhancement on MRI in approximately 50% of cases, molecular GBM frequently resemble low-grade IDH-mutant gliomas. We hypothesized that MRI-based deep learning models could distinguish IDH-mutant glioma from molecular GBM. We retrospectively identified 97 individuals with molecular GBM, who were matched to 97 individuals with IDH-mutant glioma by age, sex, year of diagnosis, and enhancement on MRI. Of the 97 IDH-mutant gliomas, 51% were 1p/19q codeleted, 42% grade 2, 47% grade 3, and 10% grade 4. Preoperative postcontrast T1-weighted (T1C) and T2-weighted (T2) MRI sequences were used to train a 3D-DenseNet121 model. The model was validated on an independent prospective set of 150 IDH-mutant gliomas and 71 molecular GBM. The MRI-based deep learning model achieved an area under the receiver operator curve (AUC) of 0.75 (95% CI: 0.68-0.82) in the prospective validation set. Sex-stratified AUCs were 0.72 (95% CI: 0.59-0.85) for females and 0.76 (95% CI: 0.68-0.85) for males. Age stratified AUCs were 0.71 (95% CI: 0.60-0.83) for patients under 55 years of age and 0.64 (95% CI: 0.50-0.79) for patients older than 55. These results suggest that discriminative MRI features of IDH-mutant glioma and molecular GBM exist and can be identified from deep learning models. However, radiographic ambiguity exists and underscores the need for further optimization of the deep learning architecture or by integrating additional non-invasive modalities. For example, future work entails integrating germline variants into the model, including rs55705857, which is a causal variant for IDH-mutant glioma. With further validation and integration of multi-modal data, a noninvasive tool may support more personalized decision-making.