Histological Grade of Endometrioid Endometrial Cancer and Relapse Risk Can Be Predicted with Machine Learning from Gene Expression Data.
Ontology highlight
ABSTRACT:
SUBMITTER: Gargya P
PROVIDER: S-EPMC8430924 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:OBJECTIVES:Stage I, grade 1 endometrial cancers have low recurrence rates and often do not receive adjuvant therapy. We compared recurrent cases to matched non-recurrent controls to evaluate for molecular markers associated with higher risk of recurrence. METHODS:A case-control study including all cases of recurrent stage I, grade 1 endometrioid endometrial cancer at one institution in a ten-year period. Cases were matched to controls by age, BMI, weight and stage. Molecular testing and immunohistochemistry were performed on archival tumor specimens: microsatellite instability (MSI-H), mismatch repair status, POLE mutational status, and next-generation sequencing. RESULTS:15 stage I, grade 1 endometrial cancer cases with recurrent disease and available tumor specimens were identified. CTNNB1 and MSI-H were present at significantly higher rates in cases than controls (CTNNB1 60% vs. 28%, OR 3.9, 95%CI 1.1-14.7, p?=?0.04 and MSI-H 53% vs. 21%, OR 4.4, 95%CI 1.1-17.0, p?=?0.03). POLE mutations were found in 0% of cases vs. 7% of controls (p?=?0.54). Among specimens demonstrating microsatellite stability (MSS), 100% of cases vs. 26% of controls had CTNNB1 mutations (p?<?0.001). CTNNB1 wild type tumors were MSI-H in 100% of cases vs. 19% of controls (p?<?0.001). CONCLUSIONS:Compared to controls, CTNNB1 mutation is present at significantly higher rates in recurrent stage I, grade 1 endometrial cancers and is found most commonly in MSS tumors. MSI-H is also present at significantly higher rates in recurrent cases. These markers may be useful for prognostic risk stratification and adjuvant therapy decision-making in this otherwise low-risk population.
| S-EPMC7005911 | biostudies-literature
Project description:Our aim was to investigate whether molecular classification can be used to refine prognosis in grade 3 endometrial endometrioid carcinomas (EECs). Grade 3 EECs were classified into 4 subgroups: p53 abnormal, based on mutant-like immunostaining (p53abn); MMR deficient, based on loss of mismatch repair protein expression (MMRd); presence of POLE exonuclease domain hotspot mutation (POLE); no specific molecular profile (NSMP), in which none of these aberrations were present. Overall survival (OS) and recurrence-free survival (RFS) rates were compared using the Kaplan-Meier method (Log-rank test) and univariable and multivariable Cox proportional hazard models. In total, 381 patients were included. The median age was 66 years (range, 33 to 96?y). Federation Internationale de Gynecologie et d'Obstetrique stages (2009) were as follows: IA, 171 (44.9%); IB, 120 (31.5%); II, 24 (6.3%); III, 50 (13.1%); IV, 11 (2.9%). There were 49 (12.9%) POLE, 79 (20.7%) p53abn, 115 (30.2%) NSMP, and 138 (36.2%) MMRd tumors. Median follow-up of patients was 6.1 years (range, 0.2 to 17.0?y). Compared to patients with NSMP, patients with POLE mutant grade 3 EEC (OS: hazard ratio [HR], 0.36 [95% confidence interval, 0.18-0.70]; P=0.003; RFS: HR, 0.17 [0.05-0.54]; P=0.003) had a significantly better prognosis; patients with p53abn tumors had a significantly worse RFS (HR, 1.73 [1.09-2.74]; P=0.021); patients with MMRd tumors showed a trend toward better RFS. Estimated 5-year OS rates were as follows: POLE 89%, MMRd 75%, NSMP 69%, p53abn 55% (Log rank P=0.001). Five-year RFS rates were as follows: POLE 96%, MMRd 77%, NSMP 64%, p53abn 47% (P=0.000001), respectively. In a multivariable Cox model that included age and Federation Internationale de Gynecologie et d'Obstetrique stage, POLE and MMRd status remained independent prognostic factors for better RFS; p53 status was an independent prognostic factor for worse RFS. Molecular classification of grade 3 EECs reveals that these tumors are a mixture of molecular subtypes of endometrial carcinoma, rather than a homogeneous group. The addition of molecular markers identifies prognostic subgroups, with potential therapeutic implications.
| S-EPMC5893364 | biostudies-literature
Project description:We have previously shown that HOMER2 (Homer scaffolding protein 2), a protein coding gene, was highly expressed in low grade (LG) endometrioid adenocarcinoma (EAC) of the uterus. The role of HOMER2 in endometrial cancer (EC) is widely unknown; therefore, the aim of this study was to determine the expression and the predictive value of HOMER2 protein expression in series of patients with EC. HOMER2 protein expression was detected on paraffin-embedded tissues from 336 cases using immunohistochemistry (IHC). Tumours were categorised in two groups; group 1 (EAC, FIGO grade 1 and 2; n = 191) and group 2 (all other subtypes including grade 3 EAC; n = 145). Statistical analysis was performed to evaluate associations between HOMER2 protein expression and pathological parameters (histological type, grade, stage, lymphovascular invasion, myometrial depth of invasion) and patient outcome [progression-free survival (PFS) and cancer-specific survival (CSS)]. HOMER2 was significantly overexpressed in group 1 compared to group 2 cancers (67% versus 30%; p < 0.001) and with low tumour grade (p < 0.001). In group 1, HOMER2 overexpression was an independent prognostic factor for improved CSS (adjusted-hazard ratio 0.28; 95% confidence interval 0.08-0.96; p = 0.042). HOMER2 expression was not associated with survival in group 2 (p > 0.05). This is the first study of HOMER2 protein expression in EC. We speculate that HOMER2 may be involved in tumourigenesis of endometrioid uterine tumours and suggest that HOMER2 should be studied further for potential clinical and therapeutic applications.
| S-EPMC7703685 | biostudies-literature
Project description:Patients with high-grade serous ovarian cancer suffer poor prognosis and variable response to treatment. Known prognostic factors for this disease include homologous recombination deficiency status, age, pathological stage and residual disease status after debulking surgery. Recent work has highlighted important prognostic information captured in computed tomography and histopathological specimens, which can be exploited through machine learning. However, little is known about the capacity of combining features from these disparate sources to improve prediction of treatment response. Here, we assembled a multimodal dataset of 444 patients with primarily late-stage high-grade serous ovarian cancer and discovered quantitative features, such as tumor nuclear size on staining with hematoxylin and eosin and omental texture on contrast-enhanced computed tomography, associated with prognosis. We found that these features contributed complementary prognostic information relative to one another and clinicogenomic features. By fusing histopathological, radiologic and clinicogenomic machine-learning models, we demonstrate a promising path toward improved risk stratification of patients with cancer through multimodal data integration.
| S-EPMC9239907 | biostudies-literature
Project description:To identify prognostic factors in patients with grade 3 (high-grade) endometrial endometrioid carcinoma, we evaluated the spectrum of genomic alterations and examined whether previously reported molecular subtypes of endometrial carcinoma were adapted to clinical outcome prediction. Seventy-five Japanese patients with grade 3 endometrial endometrioid carcinoma, who underwent a potentially curative resection procedure between 1997 and 2018 at the National Cancer Center Hospital, were included. We classified the patients into four risk groups of the disease based on the Proactive Molecular Risk Classifier for Endometrial Cancer. Genomic alterations in PTEN, ARID1A, TP53, and PIK3CA were detected in more than 30% of the patients. Overall survival and recurrence-free survival of patients with genomic alterations in CTNNB1 were poorer than those of patients with wild-type CTNNB1 (p = 0.006 and p = 0.004, respectively). Compared with that of alterations prevalent in Caucasians, the frequency of genomic alterations in POLE and TP53 was higher in our study than in The Cancer Genome Atlas dataset (p = 0.01 and p = 0.01, respectively). The tendency for recurrence-free survival in the POLE exonuclease domain mutation group was better than that in the TP53 mutation and mismatch repair-deficient groups (p = 0.08 and p = 0.07, respectively), consistent with the Proactive Molecular Risk Classifier for Endometrial Cancer risk classifier definition. The CTNNB1 mutation is a potential novel biomarker for the prognosis of patients with grade 3 endometrial endometrioid carcinoma, and prognosis classification using Proactive Molecular Risk Classifier for Endometrial Cancer may help screen Japanese patients with the disease.
| S-EPMC9128156 | biostudies-literature
Project description:Low-grade and early-stage endometrioid endometrial carcinomas (EECs) have an overall good prognosis but biomarkers identifying patients at risk of relapse are still lacking. Recently, CTNNB1 exon 3 mutation has been identified as a potential risk factor of recurrence in these patients. We evaluate the prognostic value of CTNNB1 mutation in a single-centre cohort of 218 low-grade, early-stage EECs, and the correlation with beta-catenin and LEF1 immunohistochemistry as candidate surrogate markers. CTNNB1 exon 3 hotspot mutations were evaluated by Sanger sequencing. Immunohistochemical staining of mismatch repair proteins (MLH1, PMS2, MSH2, and MSH6), p53, beta-catenin, and LEF1 was performed in representative tissue microarrays. Tumours were also reviewed for mucinous and squamous differentiation, and MELF pattern. Nineteen (8.7%) tumours harboured a mutation in CTNNB1 exon 3. Nuclear beta-catenin and LEF1 were significantly associated with CTNNB1 mutation, showing nuclear beta-catenin a better specificity and positive predictive value for CTNNB1 mutation. Tumours with CTNNB1 exon 3 mutation were associated with reduced disease-free survival (p = 0.010), but no impact on overall survival was found (p = 0.807). The risk of relapse in tumours with CTNNB1 exon 3 mutation was independent of FIGO stage, tumour grade, mismatch repair protein expression, or the presence of lymphovascular space invasion. CTNNB1 exon 3 mutation has a negative impact on disease-free survival in low-grade, early-stage EECs. Nuclear beta-catenin shows a higher positive predictive value than LEF1 for CTNNB1 exon 3 mutation in these tumours.
| S-EPMC8724178 | biostudies-literature
Project description:Gene regulatory network (GRN) inference can now take advantage of powerful machine learning algorithms to complement traditional experimental methods in building gene networks. However, the dynamical nature of embryonic development-representing the time-dependent interactions between thousands of transcription factors, signaling molecules, and effector genes-is one of the most challenging arenas for GRN prediction. In this work, we show that successful GRN predictions for a developmental network from gene expression data alone can be obtained with the Priors Enriched Absent Knowledge (PEAK) network inference algorithm. PEAK is a noise-robust method that models gene expression dynamics via ordinary differential equations and selects the best network based on information-theoretic criteria coupled with the machine learning algorithm Elastic Net. We test our GRN prediction methodology using two gene expression datasets for the purple sea urchin, Stronglyocentrotus purpuratus, and cross-check our results against existing GRN models that have been constructed and validated by over 30 years of experimental results. Our results find a remarkably high degree of sensitivity in identifying known gene interactions in the network (maximum 81.58%). We also generate novel predictions for interactions that have not yet been described, which provide a resource for researchers to use to further complete the sea urchin GRN. Published ChIPseq data and spatial co-expression analysis further support a subset of the top novel predictions. We conclude that GRN predictions that match known gene interactions can be produced using gene expression data alone from developmental time series experiments.
| S-EPMC8714117 | biostudies-literature
Project description:BackgroundThe histologic grade (HG) of breast cancer is an established prognostic factor. The grade is usually reported on a scale ranging from 1 to 3, where grade 3 tumours are the most aggressive. However, grade 2 is associated with an intermediate risk of recurrence, and carries limited information for clinical decision-making. Patients classified as grade 2 are at risk of both under- and over-treatment.MethodsRNA-sequencing analysis was conducted in a cohort of 275 women diagnosed with invasive breast cancer. Multivariate prediction models were developed to classify tumours into high and low transcriptomic grade (TG) based on gene- and isoform-level expression data from RNA-sequencing. HG2 tumours were reclassified according to the prediction model and a recurrence-free survival analysis was performed by the multivariate Cox proportional hazards regression model to assess to what extent the TG model could be used to stratify patients. The prediction model was validated in N=487 breast cancer cases from the The Cancer Genome Atlas (TCGA) data set. Differentially expressed genes and isoforms associated with HGs were analysed using linear models.ResultsThe classification of grade 1 and grade 3 tumours based on RNA-sequencing data achieved high accuracy (area under the receiver operating characteristic curve = 0.97). The association between recurrence-free survival rate and HGs was confirmed in the study population (hazard ratio of grade 3 versus 1 was 2.62 with 95 % confidence interval = 1.04-6.61). The TG model enabled us to reclassify grade 2 tumours as high TG and low TG gene or isoform grade. The risk of recurrence in the high TG group of grade 2 tumours was higher than in low TG group (hazard ratio = 2.43, 95 % confidence interval = 1.13-5.20). We found 8200 genes and 13,809 isoforms that were differentially expressed between HG1 and HG3 breast cancer tumours.ConclusionsGene- and isoform-level expression data from RNA-sequencing could be utilised to differentiate HG1 and HG3 tumours with high accuracy. We identified a large number of novel genes and isoforms associated with HG. Grade 2 tumours could be reclassified as high and low TG, which has the potential to reduce over- and under-treatment if implemented clinically.
| S-EPMC4862203 | biostudies-literature
Project description:ObjectivePatients with endometrioid endometrial cancer are stratified as high risk and low risk for extrauterine disease by surgical staging. Since patients with low-grade, minimally invasive disease do not benefit from comprehensive staging, pre-surgery stratification into a risk category may prevent unnecessary surgical staging in low risk patients. Our objective was to develop a predictive model to identify risk levels using somatic mutations that could be used preoperatively.MethodsWe classified endometrioid endometrial cancer patients in The Cancer Genome Atlas (TCGA) dataset into high risk and low risk categories: high risk patients presented with stage II, III or IV disease or stage I with high-intermediate risk features, whereas low risk patients consisted of the remaining stage I patients with either no myometrial invasion or low-intermediate risk features. Three strategies were used to build the prediction model: 1) mutational status for each gene; 2) number of somatic mutations for each gene; and 3) variant allele frequencies for each somatic mutation for each gene.ResultsEach prediction strategy had a good performance, with an area under the curve (or AUC) between 61% and 80%. Analysis of variant allele frequency produced a superior prediction model for risk levels of endometrial cancer as compared to the other two strategies, with an AUC=91%. Lasso and Ridge methods identified 53 mutations that together had the highest predictability for high risk endometrioid endometrial cancer.ConclusionsThis prediction model will assist future retrospective and prospective studies to categorize endometrial cancer patients into high risk and low risk in the preoperative setting.
| S-EPMC5257080 | biostudies-literature
Project description:Endometrial cancer (EC) incidence and mortality continues to rise. Molecular profiling of EC promises improvement of risk assessment and treatment selection. However, we still lack robust and accurate models to predict those at risk of failing treatment. The objective of this pilot study is to create models with clinical and genomic data that will discriminate patients with EC at risk of disease recurrence. We performed a pilot, retrospective, case−control study evaluating patients with EC, endometrioid type: 7 with recurrence of disease (cases), and 55 without (controls). RNA was extracted from frozen specimens and sequenced (RNAseq). Genomic features from RNAseq included transcriptome expression, genomic, and structural variation. Feature selection for variable reduction was performed with univariate ANOVA with cross-validation. Selected variables, informative for EC recurrence, were introduced in multivariate lasso regression models. Validation of models was performed in machine-learning platforms (ML) and independent datasets (TCGA). The best performing prediction models (out of >170) contained the same lncRNA features (AUC of 0.9, and 95% CI: 0.75, 1.0). Models were validated with excellent performance in ML platforms and good performance in an independent dataset. Prediction models of EC recurrence containing lncRNA features have better performance than models with clinical data alone.
| S-EPMC9785370 | biostudies-literature