Dataset Information

Sexually dimorphic computational histopathological signatures prognostic of overall survival in high-grade gliomas via deep learning.

ABSTRACT: High-grade glioma (HGG) is an aggressive brain tumor. Sex is an important factor that differentially affects survival outcomes in HGG. We used an end-to-end deep learning approach on hematoxylin and eosin (H&E) scans to (i) identify sex-specific histopathological attributes of the tumor microenvironment (TME), and (ii) create sex-specific risk profiles to prognosticate overall survival. Surgically resected H&E-stained tissue slides were analyzed in a two-stage approach using ResNet18 deep learning models, first, to segment the viable tumor regions and second, to build sex-specific prognostic models for prediction of overall survival. Our mResNet-Cox model yielded C-index (0.696, 0.736, 0.731, and 0.729) for the female cohort and C-index (0.729, 0.738, 0.724, and 0.696) for the male cohort across training and three independent validation cohorts, respectively. End-to-end deep learning approaches using routine H&E-stained slides, trained separately on male and female patients with HGG, may allow for identifying sex-specific histopathological attributes of the TME associated with survival and, ultimately, build patient-centric prognostic risk assessment models.

SUBMITTER: Verma R

PROVIDER: S-EPMC11343024 | biostudies-literature | 2024 Aug

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Sexually dimorphic computational histopathological signatures prognostic of overall survival in high-grade gliomas via deep learning.

Verma Ruchika R Alban Tyler J TJ Parthasarathy Prerana P Mokhtari Mojgan M Toro Castano Paula P Cohen Mark L ML Lathia Justin D JD Ahluwalia Manmeet M Tiwari Pallavi P

Science advances 20240823 34

High-grade glioma (HGG) is an aggressive brain tumor. Sex is an important factor that differentially affects survival outcomes in HGG. We used an end-to-end deep learning approach on hematoxylin and eosin (H&E) scans to (i) identify sex-specific histopathological attributes of the tumor microenvironment (TME), and (ii) create sex-specific risk profiles to prognosticate overall survival. Surgically resected H&E-stained tissue slides were analyzed in a two-stage approach using ResNet18 deep learni ...[more]

PMID: 39178259

Similar Datasets

Project description:BackgroundWhile molecular insights to diffuse lower-grade glioma (dLGG) have improved the basis for prognostication, most established clinical prognostic factors come from the pre-molecular era. For instance, WHO grade as a predictor for survival in dLGG with isocitrate dehydrogenase (IDH) mutation has recently been questioned. We studied the prognostic role of WHO grade in molecularly defined subgroups and evaluated earlier used prognostic factors in the current molecular setting.Material and methodsA total of 253 adults with morphological dLGG, consecutively included between 2007 and 2018, were assessed. IDH mutations, codeletion of chromosomal arms 1p/19q, and cyclin-dependent kinase inhibitor 2A/B (CDKN2A/B) deletions were analyzed.ResultsThere was no survival benefit for patients with WHO grade 2 over grade 3 IDH-mut dLGG after exclusion of tumors with known CDKN2A/B homozygous deletion (n=157) (log-rank p=0.97). This was true also after stratification for oncological postoperative treatment and when astrocytomas and oligodendrogliomas were analyzed separately. In IDH-mut astrocytomas, residual tumor volume after surgery was an independent prognostic factor for survival (HR 1.02; 95% CI 1.01-1.03; p=0.003), but not in oligodendrogliomas (HR 1.02; 95% CI 1.00-1.03; p=0.15). Preoperative tumor size was an independent predictor in both astrocytomas (HR 1.03; 95% CI 1.00-1.05; p=0.02) and oligodendrogliomas (HR 1.05; 95% CI 1.01-1.09; p=0.01). Age was not a significant prognostic factor in multivariable analyses (astrocytomas p=0.64, oligodendrogliomas p=0.08).ConclusionOur findings suggest that WHO grade is not a robust prognostic factor in molecularly well-defined dLGG. Preoperative tumor size remained a prognostic factor in both IDH-mut astrocytomas and oligodendrogliomas in our cohort, whereas residual tumor volume predicted prognosis in IDH-mut astrocytomas only. The age cutoffs for determining high risk in patients with IDH-mut dLGG from the pre-molecular era are not supported by our results.

Project description:BackgroundEnzalutamide and apalutamide are potent next-generation androgen receptor (AR) antagonists used in metastatic and non-metastatic prostate cancer. Metabolic, hormonal and immunologic effects of deep AR suppression are unknown. We hypothesized that enzalutamide and apalutamide suppress 11β-hydroxysteroid dehydrogenase-2 (11β-HSD2), which normally converts cortisol to cortisone, leading to elevated cortisol concentrations, increased ratio of active to inactive glucocorticoids and possibly suboptimal response to immunotherapy. On-treatment glucocorticoid changes might serve as an indicator of active glucocorticoid exposure and resultant adverse consequences.Patients and methodsHuman kidney tissues were stained for AR and 11β-HSD2 expression. Patients in three trials [neoadjuvant apalutamide plus leuprolide, enzalutamide ± PROSTVAC (recombinant poxvirus prostate-specific antigen vaccine) for metastatic castration-resistant prostate cancer (CRPC) and enzalutamide ± PROSTVAC for non-metastatic castration-sensitive prostate cancer] were analyzed for cortisol and its metabolites using liquid chromatography-mass spectrometry (LC-MS/MS). Progression-free survival was determined in the metastatic CRPC study of enzalutamide ± PROSTVAC for those with glucocorticoid changes above and below the median.ResultsConcurrent AR and 11β-HSD2 expression occurs only in the kidneys of men. A statistically significant rise in cortisol concentration, cortisol/cortisone ratio and tetrahydrocortisol/tetrahydrocortisone ratio with AR antagonist treatment occurred uniformly across all three trials. In the trial of enzalutamide ± PROSTVAC for metastatic CRPC, high cortisol/cortisone ratio in the enzalutamide arm was associated with significantly improved progression-free survival. However, in the enzalutamide + PROSTVAC arm, the opposite trend was observed.ConclusionEnzalutamide and apalutamide treatment toggles renal 11β-HSD2 and significantly increases indicators of and exposure to biologically active glucocorticoids, which is associated with clinical outcomes.

Project description:BackgroundConverging evidence suggests that stress alters behavioural responses in a sex-specific manner; however, the underlying molecular mechanisms of stress remain largely unknown.MethodsWe adapted unpredictable maternal separation (UMS) and adult restraint stress (RS) paradigms to mimic stress in rats in early life or adulthood, respectively. The sexual dimorphism of the prefrontal cortex was noted, and we performed RNA sequencing (RNA-Seq) to identify specific genes or pathways responsible for sexually dimorphic responses to stress. We then performed quantitative reverse transcription polymerase chain reaction (qRT-PCR) to verify the results of RNA-Seq.ResultsFemale rats exposed to either UMS or RS showed no negative effects on anxiety-like behaviours, whereas the emotional functions of the PFC were impaired markedly in stressed male rats. Leveraging differentially expressed genes (DEG) analyses, we identified sex-specific transcriptional profiles associated with stress. There were many overlapping DEGs between UMS and RS transcriptional data sets, where 1406 DEGs were associated with both biological sex and stress, while only 117 DEGs were related to stress. Notably, Uba52 and Rpl34-ps1 were the first-ranked hub gene in 1406 and 117 DEGs respectively, and Uba52 was higher than Rp134-ps1, suggesting that stress may have led to a more pronounced effect on the set of 1406 DEGs. Pathway analysis revealed that 1406 DEGs were primarily enriched in ribosomal pathway. These results were confirmed by qRT-PCR.LimitationsSex-specific transcriptional profiles associated with stress were identified in this study, but more in-depth experiments, such as single-cell sequencing and manipulation of male and female gene networks in vivo, are needed to verify our findings.ConclusionOur findings show sex-specific behavioural responses to stress and highlight sexual dimorphism at the transcriptional level, shedding light on developing sex-specific therapeutic strategies for stress-related psychiatric disorders.

Dataset Information

Sexually dimorphic computational histopathological signatures prognostic of overall survival in high-grade gliomas via deep learning.

Publications

Sexually dimorphic computational histopathological signatures prognostic of overall survival in high-grade gliomas via deep learning.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets