Ilona Urbarova
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Ilona Urbarova has a total of 137 co-authors. Here are the top co-authors by number of shared publications:
- Therese Haugdahl Nost
Department of Community Medicine, Faculty of Health Sciences, UiT The Arctic University of Norway, Tromsø, Norway
12 shared publications - Hilde Langseth
Department of Research, Cancer Registry of Norway, Norwegian Institute of Public Health, Oslo, Norway
10 shared publications - Ioannis Fotopoulos
Department of Computer Science, University of Crete, Heraklion, Greece
10 shared publications - Luca Pestarino
Department of Research, Cancer Registry of Norway, Norwegian Institute of Public Health, Oslo, Norway
10 shared publications - Oluf Dimitri Røe
Department of Clinical and Molecular Medicine, NTNU-Norwegian University of Science and Technology, Trondheim, Norway
10 shared publications - Renee Turzanski Fortner
Division of Cancer Epidemiology German Cancer Research Center (DKFZ) Heidelberg Germany
10 shared publications
Latest Publications
- Open Access
- 18 October 2026
- Published by Elsevier in Brain, Behavior, and Immunity
- Vol. 140
Educational attainment and systemic inflammation: Evidence of a stable gradient from two measurements in the Tromsø Study (2007–2016)
- 18 October 2026
- Published by Elsevier in Brain, Behavior, and Immunity
- Vol. 140
Open AccessEducational attainment and systemic inflammation: Evidence of a stable gradient from two measurements in the Tromsø Study (2007–2016)
- Open Access
- 20 May 2026
- Published by Springer Nature in Clinical Epigenetics
- Vol. 18 (1)
Association between pace of aging estimated from blood DNA methylation and all-cause mortality: the HUNT study
Background: Epigenetic clocks, developed using blood DNA methylation data, can be used to estimate biological ages and pace of aging. We aimed to identify potential determinants of the pace of aging, estimated using blood DNA methylation, and to investigate the association between the pace of aging and all-cause mortality in a population-based Norwegian cohort with repeated DNA methylation measurements. Methods: This study included 140 cancer-free controls from a lung cancer nested case-control study within the Trøndelag Health Study (HUNT). DNA methylation was measured in blood samples in both HUNT2 (1995–97) and HUNT3 (2006–08), 11 years apart. The pace of aging was estimated using two established measures of biological age, DNAmPhenoAge and DNAmGrimAge2, and two direct measures of pace of aging, DunedinPoAm and DunedinPACE, all derived from blood DNA methylation data. All-cause mortality was followed up until 2023. Results: There was a moderate to excellent reliability of the repeated measurements, with intraclass correlation coefficient (ICC) values ranged from 0.69–0.91. University education appeared to be associated with a slower pace of aging, while smoking and obesity were associated with a faster pace. A one-standard deviation increase in the pace of aging, as measured by DNAmGrimAge2, was associated with a hazard ratio (HR) of 2.42 (95% confidence interval (CI) 1.25 to 4.68) for all-cause mortality in HUNT2, and a HR of 2.30 (95% CI 1.22 to 4.33) in HUNT3 after adjustment for the established risk factors. Conclusions: The pace of aging, as estimated using blood DNA methylation, appears to be an independent predictor of all-cause mortality. This measure may reflect the combined influence of genetic, lifestyle, and environmental factors on individual aging trajectories.
- 20 May 2026
- Published by Springer Nature in Clinical Epigenetics
- Vol. 18 (1)
Open AccessAssociation between pace of aging estimated from blood DNA methylation and all-cause mortality: the HUNT study
Background: Epigenetic clocks, developed using blood DNA methylation data, can be used to estimate biological ages and pace of aging. We aimed to identify potential determinants of the pace of aging, estimated using blood DNA methylation, and to investigate the association between the pace of aging and all-cause mortality in a population-based Norwegian cohort with repeated DNA methylation measurements. Methods: This study included 140 cancer-free controls from a lung cancer nested case-control study within the Trøndelag Health Study (HUNT). DNA methylation was measured in blood samples in both HUNT2 (1995–97) and HUNT3 (2006–08), 11 years apart. The pace of aging was estimated using two established measures of biological age, DNAmPhenoAge and DNAmGrimAge2, and two direct measures of pace of aging, DunedinPoAm and DunedinPACE, all derived from blood DNA methylation data. All-cause mortality was followed up until 2023. Results: There was a moderate to excellent reliability of the repeated measurements, with intraclass correlation coefficient (ICC) values ranged from 0.69–0.91. University education appeared to be associated with a slower pace of aging, while smoking and obesity were associated with a faster pace. A one-standard deviation increase in the pace of aging, as measured by DNAmGrimAge2, was associated with a hazard ratio (HR) of 2.42 (95% confidence interval (CI) 1.25 to 4.68) for all-cause mortality in HUNT2, and a HR of 2.30 (95% CI 1.22 to 4.33) in HUNT3 after adjustment for the established risk factors. Conclusions: The pace of aging, as estimated using blood DNA methylation, appears to be an independent predictor of all-cause mortality. This measure may reflect the combined influence of genetic, lifestyle, and environmental factors on individual aging trajectories.
- Open Access
- 3 October 2025
- Published by American Association for Cancer Research (AACR)
Supplementary Table 2 from Recommendations for Validating Omics Prediction Models: Insights from a Lung Cancer RNA Biomarker Study
This file contains the list of RNA features used in the XGBoost classification models. Each RNA class (All RNAs, miRNA, miscRNA) is shown in a separate sheet. For each RNA, its relative importance is reported for each histology class evaluated.
- 3 October 2025
- Published by American Association for Cancer Research (AACR)
Open AccessSupplementary Table 2 from Recommendations for Validating Omics Prediction Models: Insights from a Lung Cancer RNA Biomarker Study
This file contains the list of RNA features used in the XGBoost classification models. Each RNA class (All RNAs, miRNA, miscRNA) is shown in a separate sheet. For each RNA, its relative importance is reported for each histology class evaluated.
- Open Access
- 3 October 2025
- Published by American Association for Cancer Research (AACR)
Data from Recommendations for Validating Omics Prediction Models: Insights from a Lung Cancer RNA Biomarker Study
Background:External validation of predictive models in medical research is crucial to ensure their generalizability and applicability across diverse populations. However, validation often reveals discrepancies in model performance due to cohort differences, sample collection and storage, overfitting, and inconsistencies in data handling. This study investigates the challenges encountered during external validation of predictive models for early lung cancer detection using small RNA biomarkers, tying these challenges to specific validation outcomes and deriving recommendations.Methods:Predictive models based on the XGBoost algorithm, developed from serum samples in the JanusRNA cohort, were externally validated in two independent Norwegian cohorts: HUNT and Norwegian Women and Cancer Study. These cohorts differed in sample types, RNA abundance, library preparation protocols, and lung cancer histologic classification. Strategies to harmonize data processing and address these discrepancies were used to ensure a robust validation process.Results:Validation revealed significant challenges due to cohort heterogeneity. Median AUC values ranged from 0.50 to 0.66 in validation cohorts, compared with 0.62 to 0.76 in the original models. Models performed worse in the female-only Norwegian Women and Cancer Study cohort, in which plasma was used, highlighting the impact of sample type and cohort characteristics on predictive accuracy.Conclusion:Based on the challenges encountered during validation, we propose seven recommendations to guide robust external validation of omics-based predictive models, including harmonizing data processing across cohorts, reevaluating overfitting, and critically assessing model performance for clinical applications.Impact:By highlighting practical issues in model validation and providing recommendations, this study supports more reliable and clinically applicable biomarker-based prediction models, ultimately aiding cancer screening and prevention efforts.
- 3 October 2025
- Published by American Association for Cancer Research (AACR)
Open AccessData from Recommendations for Validating Omics Prediction Models: Insights from a Lung Cancer RNA Biomarker Study
Background:External validation of predictive models in medical research is crucial to ensure their generalizability and applicability across diverse populations. However, validation often reveals discrepancies in model performance due to cohort differences, sample collection and storage, overfitting, and inconsistencies in data handling. This study investigates the challenges encountered during external validation of predictive models for early lung cancer detection using small RNA biomarkers, tying these challenges to specific validation outcomes and deriving recommendations.Methods:Predictive models based on the XGBoost algorithm, developed from serum samples in the JanusRNA cohort, were externally validated in two independent Norwegian cohorts: HUNT and Norwegian Women and Cancer Study. These cohorts differed in sample types, RNA abundance, library preparation protocols, and lung cancer histologic classification. Strategies to harmonize data processing and address these discrepancies were used to ensure a robust validation process.Results:Validation revealed significant challenges due to cohort heterogeneity. Median AUC values ranged from 0.50 to 0.66 in validation cohorts, compared with 0.62 to 0.76 in the original models. Models performed worse in the female-only Norwegian Women and Cancer Study cohort, in which plasma was used, highlighting the impact of sample type and cohort characteristics on predictive accuracy.Conclusion:Based on the challenges encountered during validation, we propose seven recommendations to guide robust external validation of omics-based predictive models, including harmonizing data processing across cohorts, reevaluating overfitting, and critically assessing model performance for clinical applications.Impact:By highlighting practical issues in model validation and providing recommendations, this study supports more reliable and clinically applicable biomarker-based prediction models, ultimately aiding cancer screening and prevention efforts.
- Open Access
- 3 October 2025
- Published by American Association for Cancer Research (AACR)
Supplementary Table 4 from Recommendations for Validating Omics Prediction Models: Insights from a Lung Cancer RNA Biomarker Study
Performance metrics (AUC, sensitivity at 0.95 specificity, accuracy, recall, and precision) with 95% confidence intervals across the different cohorts for each histology and RNA type. Results are averaged over the five train and test splits.
- 3 October 2025
- Published by American Association for Cancer Research (AACR)
Open AccessSupplementary Table 4 from Recommendations for Validating Omics Prediction Models: Insights from a Lung Cancer RNA Biomarker Study
Performance metrics (AUC, sensitivity at 0.95 specificity, accuracy, recall, and precision) with 95% confidence intervals across the different cohorts for each histology and RNA type. Results are averaged over the five train and test splits.
- Open Access
- 3 October 2025
- Published by American Association for Cancer Research (AACR)
Supplementary Figure 4 from Recommendations for Validating Omics Prediction Models: Insights from a Lung Cancer RNA Biomarker Study
Validation AUC values across the JanusRNA versions, RNA type and histologies. Each boxplot represents AUC values for individual test set repetitions, categorized by RNA types and histological groups. 'JanusRNA - LC (Umu and colleagues, 2022)' refers to the AUC values for the JanusRNA lung cancer data as described by Umu and colleagues, while ‘JanusRNA - LC (updated)’ shows the values in the JanusRNA lung cancer samples after transitioning to the sncRNA-workflow v.1.01 pipeline. ‘JanusRNA - LC (updated) Females' indicates the JanusRNA lung cancer samples with updated pipeline and including only female samples.
- 3 October 2025
- Published by American Association for Cancer Research (AACR)
Open AccessSupplementary Figure 4 from Recommendations for Validating Omics Prediction Models: Insights from a Lung Cancer RNA Biomarker Study
Validation AUC values across the JanusRNA versions, RNA type and histologies. Each boxplot represents AUC values for individual test set repetitions, categorized by RNA types and histological groups. 'JanusRNA - LC (Umu and colleagues, 2022)' refers to the AUC values for the JanusRNA lung cancer data as described by Umu and colleagues, while ‘JanusRNA - LC (updated)’ shows the values in the JanusRNA lung cancer samples after transitioning to the sncRNA-workflow v.1.01 pipeline. ‘JanusRNA - LC (updated) Females' indicates the JanusRNA lung cancer samples with updated pipeline and including only female samples.
- Open Access
- 3 October 2025
- Published by American Association for Cancer Research (AACR)
Supplementary Figure 2 from Recommendations for Validating Omics Prediction Models: Insights from a Lung Cancer RNA Biomarker Study
Comparison of average sequence distribution length across the three versions of the JanusRNA data (panel A) and RNA type distribution (panel B).
- 3 October 2025
- Published by American Association for Cancer Research (AACR)
Open AccessSupplementary Figure 2 from Recommendations for Validating Omics Prediction Models: Insights from a Lung Cancer RNA Biomarker Study
Comparison of average sequence distribution length across the three versions of the JanusRNA data (panel A) and RNA type distribution (panel B).
- Open Access
- 3 October 2025
- Published by American Association for Cancer Research (AACR)
Supplementary Figure 3 from Recommendations for Validating Omics Prediction Models: Insights from a Lung Cancer RNA Biomarker Study
Comparison of total reads distribution across the different versions of the JanusRNA data.
- 3 October 2025
- Published by American Association for Cancer Research (AACR)
Open AccessSupplementary Figure 3 from Recommendations for Validating Omics Prediction Models: Insights from a Lung Cancer RNA Biomarker Study
Comparison of total reads distribution across the different versions of the JanusRNA data.
- Open Access
- 3 October 2025
- Published by American Association for Cancer Research (AACR)
Supplementary Table 3 from Recommendations for Validating Omics Prediction Models: Insights from a Lung Cancer RNA Biomarker Study
Cohort characteristics of the original JanusRNA - LC cohort from Umu and colleagues, 2022, compared with those on the same cohort after applying the updated bioinformatics pipeline and after including female samples only.
- 3 October 2025
- Published by American Association for Cancer Research (AACR)
Open AccessSupplementary Table 3 from Recommendations for Validating Omics Prediction Models: Insights from a Lung Cancer RNA Biomarker Study
Cohort characteristics of the original JanusRNA - LC cohort from Umu and colleagues, 2022, compared with those on the same cohort after applying the updated bioinformatics pipeline and after including female samples only.
- Open Access
- 3 October 2025
- Published by American Association for Cancer Research (AACR)
Supplementary Figure 1 from Recommendations for Validating Omics Prediction Models: Insights from a Lung Cancer RNA Biomarker Study
Comparison of total reads distribution across the three cohorts: JanusRNA - LC (updated), HUNT and NOWAC.
- 3 October 2025
- Published by American Association for Cancer Research (AACR)
Open AccessSupplementary Figure 1 from Recommendations for Validating Omics Prediction Models: Insights from a Lung Cancer RNA Biomarker Study
Comparison of total reads distribution across the three cohorts: JanusRNA - LC (updated), HUNT and NOWAC.