Predictive radiogenomics modeling of EGFR mutation status in lung cancer.
Ontology highlight
ABSTRACT:
SUBMITTER: Gevaert O
PROVIDER: S-EPMC5282551 | biostudies-literature | 2017 Jan
REPOSITORIES: biostudies-literature
Publications
Predictive radiogenomics modeling of EGFR mutation status in lung cancer.
Gevaert Olivier O Echegaray Sebastian S Khuong Amanda A Hoang Chuong D CD Shrager Joseph B JB Jensen Kirstin C KC Berry Gerald J GJ Guo H Henry HH Lau Charles C Plevritis Sylvia K SK Rubin Daniel L DL Napel Sandy S Leung Ann N AN
Scientific reports 20170131
Molecular analysis of the mutation status for EGFR and KRAS are now routine in the management of non-small cell lung cancer. Radiogenomics, the linking of medical images with the genomic properties of human tumors, provides exciting opportunities for non-invasive diagnostics and prognostics. We investigated whether EGFR and KRAS mutation status can be predicted using imaging data. To accomplish this, we studied 186 cases of NSCLC with preoperative thin-slice CT scans. A thoracic radiologist anno ...[more]
Similar Datasets
Project description:Background Lung cancer is the most common primary tumor metastasizing to the brain. A significant proportion of lung cancer patients show epidermal growth factor receptor (EGFR) mutation status discordance between the primary cancer and the corresponding brain metastases, which can affect prognosis and therapeutic decision-making. However, it is not always feasible to obtain brain metastases samples. The aim of this study was to establish a radiomic model to predict the EGFR mutation status of lung cancer brain metastases. Methods Data from 162 patients with resected brain metastases originating from lung cancer (70 with mutant EGFR, 92 with wild-type EGFR) were retrospectively analyzed. Radiomic features were extracted using preoperative brain magnetic resonance (MR) images (contrast-enhanced T1-weighted imaging, T1CE; T2-weighted imaging, T2WI; T2 fluid-attenuated inversion recovery, T2 FLAIR; and combinations of these sequences), to establish machine learning-based models for predicting the EGFR status of excised brain metastases (108 metastases for training and 54 metastases for testing). The least absolute shrinkage selection operator was used to select informative features; radiomics models were built with logistic regression of the training cohort, and model performance was evaluated using an independent test set. Results The best-performing model was a combination of 10 features selected from multiple sequences (two from T1CE, five from T2WI, and three from T2 FLAIR) in both the training and test sets, resulting in classification area under the curve, accuracy, sensitivity, and specificity values of 0.85 and 0.81, 77.8% and 75.9%, 83.7% and 73.1%, and 73.8% and 78.6%, respectively. Conclusions Radiomic signatures integrating multi-sequence MR images have the potential to noninvasively predict the EGFR mutation status of lung cancer brain metastases.
| S-EPMC9334014 | biostudies-literature
Project description:Radiogenomics uses machine-learning (ML) to directly connect the morphologic and physiological appearance of tumors on clinical imaging with underlying genomic features. Despite extensive growth in the area of radiogenomics across many cancers, and its potential role in advancing clinical decision making, no published studies have directly addressed uncertainty in these model predictions. We developed a radiogenomics ML model to quantify uncertainty using transductive Gaussian Processes (GP) and a unique dataset of 95 image-localized biopsies with spatially matched MRI from 25 untreated Glioblastoma (GBM) patients. The model generated predictions for regional EGFR amplification status (a common and important target in GBM) to resolve the intratumoral genetic heterogeneity across each individual tumor-a key factor for future personalized therapeutic paradigms. The model used probability distributions for each sample prediction to quantify uncertainty, and used transductive learning to reduce the overall uncertainty. We compared predictive accuracy and uncertainty of the transductive learning GP model against a standard GP model using leave-one-patient-out cross validation. Additionally, we used a separate dataset containing 24 image-localized biopsies from 7 high-grade glioma patients to validate the model. Predictive uncertainty informed the likelihood of achieving an accurate sample prediction. When stratifying predictions based on uncertainty, we observed substantially higher performance in the group cohort (75% accuracy, n = 95) and amongst sample predictions with the lowest uncertainty (83% accuracy, n = 72) compared to predictions with higher uncertainty (48% accuracy, n = 23), due largely to data interpolation (rather than extrapolation). On the separate validation set, our model achieved 78% accuracy amongst the sample predictions with lowest uncertainty. We present a novel approach to quantify radiogenomics uncertainty to enhance model performance and clinical interpretability. This should help integrate more reliable radiogenomics models for improved medical decision-making.
| S-EPMC7886858 | biostudies-literature
Project description:Patients with advanced non-small-cell lung cancer (NSCLC) and somatic activating mutations of the tyrosine kinase (TK) domain of the epidermal growth factor receptor (EGFR) gene represent a biologically distinct disease entity that shows exquisite sensitivity to the reversible EGFR-TK inhibitors (-TKIs) gefitinib or erlotinib. Phase III randomized studies have clearly demonstrated that a reversible EGFR-TKI is significantly superior in terms of response rate, progression-free survival and quality of life to platinum-based chemotherapy in advanced NSCLC patients who carry an activating EGFR mutation, thus resulting into a new standard of care for this biologically selected group of patients. Unfortunately, approximately one third of EGFR-mutated patients show primary resistance to gefitinib or erlotinib, whereas virtually all patients who initially benefit from treatment will eventually develop acquired resistance. Importantly, revealing the molecular mechanisms that underlie resistance to reversible EGFR-TKIs is key to the development of EGFR-targeting strategies with the potential to prevent, delay or overcome such resistance. Early results of clinical trials with irreversible EGFR-TKIs or dual combination strategies aiming to block EGFR-mediated signaling at different levels have shown encouraging results in EGFR-mutated patients pretreated or not with a reversible EGFR-TKI. Therefore, in the near future it is reasonable to hypothesize that EGFR-mutated NSCLCs could be treated with multiple lines of EGFR-targeting therapies beyond disease progression, limiting chemotherapy to selected cases of resistant disease. This evolving treatment scenario highlights once again how important is the identification of a single oncogenic "addiction" that functions as unique determinant of progression and survival of NSCLC.
| S-EPMC4367583 | biostudies-other
Project description:Epidermal growth factor receptor (EGFR) mutations are potential markers driving carcinogenesis, and may alter the response to EGFR tyrosine kinase inhibitors in patients with non-small cell lung cancer (NSCLC). The frequency of EGFR mutations in patients with NSCLC differs according to sex, smoking habits and regional-based ethnicity differences. The aim of the present study was to determine the frequency of EGFR mutations in Turkish patients with NSCLC to highlight the importance of regional differences, and their associations with patient characteristics. Genomic DNA was extracted from formalin-fixed and paraffin-embedded tumor tissue sections of 409 NSCLC patients. The most common EGFR mutations in exons 18, 19, 20 and 21 were detected using BioFilmChip-based microarray assay. The overall EGFR mutation frequency was 16.6%, and the highest mutation frequencies were observed in exon 19 (6.4%) and exon 21 (7.3%). There was a higher frequency of EGFR mutations in females compared with males and in never-smokers compared with smokers (both P?0.05). These results were similar to other European population-based studies, but not consistent Middle-Eastern based studies. The present study may contribute to understanding the gradient frequency of EGFR mutation across different ethnicities, and in designing genome wide-based collaborations that may reveal novel decision making and susceptibility mutations in EGFR in patients with NSCLC.
| S-EPMC7271734 | biostudies-literature
Project description:Background and objectiveEpidermal growth factor receptor (EGFR) mutation status is used as a predictive biomarker for the tyrosine kinase inhibitors therapy in non-small cell lung cancer (NSCLC). The incidence of EGFR mutations appears to vary according to ethnic and geographical backgrounds. This retrospective study aimed to investigate the EGFR mutation status in Algerian NSCLC patients and its association with clinicopathological features.MethodsWe examined the presence of EGFR mutations (Exons 19-21) in 58 unselected NSCLC samples using PCR followed by direct sequencing.ResultsThe present study included 53 (91.4%) men and 5 (8.6%) women, with a median age of 59 (ranging from 44 to 94 years old). EGFR mutations were detected in 23 patients, with an overall rate of 39.6%. There were 21 (91.3%) cases with the exon-21 L585R single mutation and two (8.7%) with dual mutations of exon-19 deletions and L585R. EGFR mutations were more frequently found in patients with confirmed adenocarcinoma (14/27, 51.8%) than in non-adenomatous NCSCL subtypes (3/14, 21.4%; p=0.03). Furthermore, early stages of the disease were significantly associated with a higher rate of EGFR mutations (14/27, 51.8%) compared with those at advanced stage (5/21, 23.8%; p=0.02). There were no significant differences in EGFR mutation frequency by age, gender, or smoking status.ConclusionWe found that Algerian NSCLC patients exhibited a high rate of EGFR mutations, which was quite similar to that in Asians population rather than Caucasian patients. Thus, TKI-based treatments may be more beneficial for Algerian patients with NSCLC. Further studies using a large number of patients are required to confirm our preliminary findings.
| S-EPMC8325123 | biostudies-literature
Project description:Epidermal growth factor receptor (EGFR) mutations in non-small cell lung cancer (NSCLC) are a response to EGFR-tyrosine kinase inhibitor. However, a lack of sufficient tumor tissue has been a limitation for determining EGFR mutation status in clinical practice. The objective of this study was to predict EGFR mutation status in NSCLC patients based on a model including maximum standardized uptake value (SUVmax) and clinical features.We retrospectively reviewed NSCLC patients undergoing EGFR mutation testing and pretreatment positron emission tomography/computed tomography between March 2009 and December 2013. The relationships of EGFR mutations with both SUVmax and patient characteristics were evaluated, and a multivariate logistic regression analysis was performed. The model was assessed by area under the receiver-operating characteristic curve (AUC) and was prospectively validated during January to June 2014.Three hundred and sixteen patients meeting the criteria were enrolled for model construction. The SUVmax values were significantly lower for EGFR mutations (mean, 9.5 ± 5.74) than for EGFR wild-type (mean, 12.7 ± 6.43; P < 0.001). ROC curve analysis showed that the SUVmax cutoff point was 8.1, for which the AUC was 0.65 (95% confidence interval [CI], 0.60-0.72). In addition, multivariate analysis also showed that low SUVmax (≤8.1) was a predictor of EGFR mutations, for which the AUC was 0.77, combining nonsmoking history and primary tumor size (≤5 cm). Eighty-five patients were enrolled to validate the predictive model, and the overall accuracy, sensitivity, and specificity were 77.6%, 64.6% (95% CI 40.7-82.8), and 82.5% (95% CI 70.9-91.0), respectively.The specific FDG uptake value could be considered to effectively predict EGFR mutation status of NSCLC patients by considering smoking history and primary tumor size when genetic tests are not available.
| S-EPMC5265876 | biostudies-literature
Project description:EGFR and KRAS are the most frequently mutated genes in lung cancer, being active research topics in targeted therapy. The biopsy is the traditional method to genetically characterise a tumour. However, it is a risky procedure, painful for the patient, and, occasionally, the tumour might be inaccessible. This work aims to study and debate the nature of the relationships between imaging phenotypes and lung cancer-related mutation status. Until now, the literature has failed to point to new research directions, mainly consisting of results-oriented works in a field where there is still not enough available data to train clinically viable models. We intend to open a discussion about critical points and to present new possibilities for future radiogenomics studies. We conducted high-dimensional data visualisation and developed classifiers, which allowed us to analyse the results for EGFR and KRAS biological markers according to different combinations of input features. We show that EGFR mutation status might be correlated to CT scans imaging phenotypes; however, the same does not seem to hold for KRAS mutation status. Also, the experiments suggest that the best way to approach this problem is by combining nodule-related features with features from other lung structures.
| S-EPMC7046701 | biostudies-literature
Project description:BackgroundIn this study we retrospectively evaluated the capability of computed tomography (CT)-based radiomic features to predict epidermal growth factor receptor (EGFR) mutation status in surgically-resected peripheral lung adenocarcinomas in an Asian cohort of patients.Patients and methodsTwo hundred ninety-eight patients with surgically resected peripheral lung adenocarcinomas were investigated in this institutional review board-approved retrospective study with requirement waived to obtain informed consent. Two hundred nineteen quantitative 3-D features were extracted from segmented volumes of each tumor, and 59 of these, which were considered independent features, were included in the analysis. Clinical and pathological information was obtained from the institutional database.ResultsMutant EGFR was significantly associated with female sex (P = .0005); never smoker status (P < .0001), lepidic predominant adenocarcinomas (P = .017), and low or intermediate pathologic grade (P = .0002). Statistically significant differences were found in 11 radiomic features between EGFR mutant and wild type groups in univariate analysis. Mutant EGFR status could be predicted by a set of 5 radiomic features that fell into 3 broad groups: CT attenuation energy, tumor main direction, and texture defined according to wavelets and Laws (area under the curve [AUC], 0.647). A multiple logistic regression model showed that adding radiomic features to a clinical model resulted in a significant improvement of predicting power, because the AUC increased from 0.667 to 0.709 (P < .0001).ConclusionComputed tomography-based radiomic features of peripheral lung adenocarcinomas can capture useful information regarding tumor phenotype, and the model we built can be useful to predict the presence of EGFR mutations in peripheral lung adenocarcinoma in Asian patients when mutational profiling is not available or possible.
| S-EPMC5548419 | biostudies-literature
Project description:BackgroundEpidermal growth factor receptor (EGFR) mutation status is the most valuable indicator in the screening of non-small-cell lung cancer (NSCLC) patients for tyrosine kinase inhibitor (TKI) therapy. Accurate, rapid and economical methods of detecting EGFR mutations have become important. The use of two mutation-specific antibodies targeting the delE746-A750 mutation in exon 19 and L858R mutation in exon 21 makes this task possible, but the lack of consensually acceptable criteria for positive results limits the application of this antibody based mutation detection.MethodsWe collected 399 specimens from NSCLC patients (145 resection specimens, 220 biopsy specimens, and 34 cytology specimens) whose EGFR mutation status had been detected by TaqMan PCR assay. Immunohistochemical (IHC) analyses using EGFR mutation-specific antibodies were employed for all samples. After staining and scoring, the sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) were calculated in accordance with different levels of positive grades in comparison with the results of PCR-based assay.ResultsIn IHC-based analyses, 144 cases were scored 0, 104 cases were scored 1+, 103 cases were scored 2+, and 48 cases were scored 3+. With the molecular-based results were set as the "gold standard", the prevalence of mutation was 6.94% (10/144), 23.08% (24/104), 67.96% (70/103) and 100% (48/48), respectively, for samples with scores 0, 1+, 2+ and 3+. When score 3+ was considered positive, the specificity and PPV were 100%; if only score 0 was considered negative, 93.06% NPV was obtained.ConclusionPatients with score 3+ have a perfect PPV (100%), and may accept TKI treatment directly without any molecular-based assays. Patients with score 0 had high NPV (93.06%), which could reach 97.22% when the detection of total EGFR was applied. However, samples with score 1+ or 2+ are unreliable and need further verification of EGFR mutation status by molecular-based assays.
| S-EPMC3594188 | biostudies-literature
Project description:AIMS:EGFR mutations now guide the clinical use of EGFR-targeted therapy in lung cancer. However, standard EGFR mutation analysis requires a minimum amount of tumor tissue, which may not be available in certain situations. In this study, we combined a mass spectrometry genotyping assay (Sequenom) with a mutant-enriched PCR (ME-PCR) to detect EGFR mutations in free plasma DNA from patients with lung cancer. METHOD:DNAs were extracted from 31 plasma samples from 31 patients and analyzed by both methods for EGFR Exon 19 deletion and EGFR L858R mutation. Results in plasma DNA samples were compared with EGFR mutation status obtained in tumor DNA (18/31 EGFR mutant). The relationship of EGFR mutation status in tumor and/or plasma samples to overall survival was assessed. RESULTS:The EGFR mutation status in plasma DNA was identical to the primary tumor in 61% of patients (19/31). By mass spectrometry genotyping, the plasma samples contained mutant DNA corresponding to 5/14 EGFR Exon 19 deletions and 3/4 EGFR L858R mutations previously diagnosed in the matched tumors. Two samples were positive in plasma DNA but negative in primary tumor tissue. Results were similar for samples studied by ME-PCR. For patients treated with erlotinib, overall survival was correlated with the presence of EGFR mutation in plasma and/or tumor tissue (p=0.002), with the two patients positive only in plasma DNA showing responses and favorable outcomes. CONCLUSION:The detection of EGFR mutations in plasma DNA samples by mass spectrometry genotyping and ME-PCR is feasible. A positive EGFR result in plasma DNA has a high predictive value for tumor EGFR status and for favorable clinical course on EGFR-targeted therapy and could therefore be useful in guiding clinical decisions in patients with insufficient or unavailable tumor specimens.
| S-EPMC3282180 | biostudies-literature