Dataset Information

Prediction of chronological and biological age from laboratory data.

ABSTRACT: Aging has pronounced effects on blood laboratory biomarkers used in the clinic. Prior studies have largely investigated one biomarker or population at a time, limiting a comprehensive view of biomarker variation and aging across different populations. Here we develop a supervised machine learning approach to study aging using 356 blood biomarkers measured in 67,563 individuals across diverse populations. Our model predicts age with a mean absolute error (MAE), or average magnitude of prediction errors, in held-out data of 4.76 years and an R2 value of 0.92. Age prediction was highly accurate for the pediatric cohort (MAE = 0.87, R2 = 0.94) but inaccurate for ages 65+ (MAE = 4.30, R2 = 0.25). Variability was observed in which biomarkers carry predictive power across age groups, genders, and race/ethnicity groups, and novel candidate biomarkers of aging were identified for specific age ranges (e.g. Vitamin E, ages 18-44). We show that predictors for one age group may fail to generalize to other groups and investigate non-linearity in biomarkers near adulthood. As populations worldwide undergo major demographic changes, it is increasingly important to catalogue biomarker variation across age groups and discover new biomarkers to distinguish chronological and biological aging.

SUBMITTER: Sagers L

PROVIDER: S-EPMC7244024 | biostudies-literature | 2020 May

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Prediction of chronological and biological age from laboratory data.

Sagers Luke L Melas-Kyriazi Luke L Patel Chirag J CJ Manrai Arjun K AK

Aging 20200505 9

Aging has pronounced effects on blood laboratory biomarkers used in the clinic. Prior studies have largely investigated one biomarker or population at a time, limiting a comprehensive view of biomarker variation and aging across different populations. Here we develop a supervised machine learning approach to study aging using 356 blood biomarkers measured in 67,563 individuals across diverse populations. Our model predicts age with a mean absolute error (MAE), or average magnitude of prediction ...[more]

PMID: 32391803

Similar Datasets

Project description:BackgroundEpigenetic clocks can track both chronological age (cAge) and biological age (bAge). The latter is typically defined by physiological biomarkers and risk of adverse health outcomes, including all-cause mortality. As cohort sample sizes increase, estimates of cAge and bAge become more precise. Here, we aim to develop accurate epigenetic predictors of cAge and bAge, whilst improving our understanding of their epigenomic architecture.MethodsFirst, we perform large-scale (N = 18,413) epigenome-wide association studies (EWAS) of chronological age and all-cause mortality. Next, to create a cAge predictor, we use methylation data from 24,674 participants from the Generation Scotland study, the Lothian Birth Cohorts (LBC) of 1921 and 1936, and 8 other cohorts with publicly available data. In addition, we train a predictor of time to all-cause mortality as a proxy for bAge using the Generation Scotland cohort (1214 observed deaths). For this purpose, we use epigenetic surrogates (EpiScores) for 109 plasma proteins and the 8 component parts of GrimAge, one of the current best epigenetic predictors of survival. We test this bAge predictor in four external cohorts (LBC1921, LBC1936, the Framingham Heart Study and the Women's Health Initiative study).ResultsThrough the inclusion of linear and non-linear age-CpG associations from the EWAS, feature pre-selection in advance of elastic net regression, and a leave-one-cohort-out (LOCO) cross-validation framework, we obtain cAge prediction with a median absolute error equal to 2.3 years. Our bAge predictor was found to slightly outperform GrimAge in terms of the strength of its association to survival (HRGrimAge = 1.47 [1.40, 1.54] with p = 1.08 × 10-52, and HRbAge = 1.52 [1.44, 1.59] with p = 2.20 × 10-60). Finally, we introduce MethylBrowsR, an online tool to visualise epigenome-wide CpG-age associations.ConclusionsThe integration of multiple large datasets, EpiScores, non-linear DNAm effects, and new approaches to feature selection has facilitated improvements to the blood-based epigenetic prediction of biological and chronological age.

Project description:BACKGROUND:The number of nonagenarian cancer patients (? 90?years) is continuously increasing, and radiotherapy is performed in a relevant proportion of patients, as surgery and chemotherapy are often not feasible for these patients. However, the evidence regarding the feasibility and treatment outcomes after radiotherapy for this patient group is very limited. METHODS:All nonagenarian patients receiving (chemo) radiotherapy between 2009 and 2019 at the University of Freiburg - Medical Center were analyzed for patterns of care, overall survival (OS) and therapy-associated toxicities according to the Common Terminology Criteria for Adverse Events. Uni- and multivariate Cox regression analyses were conducted to assess the influence of patient- and treatment-related factors on patient outcomes. RESULTS:One hundred nineteen patients with a total of 137 irradiated lesions were included in this analysis. After a median follow-up of 27?months, median OS was 10?months with a 3-year OS amounting to 11.1%. Univariate analyses demonstrated that a reduced performance status (HR?=?1.56, 95% CI 1.00-2.45, p?<?0.05), a higher burden of comorbidities (HR?=?2.00, 95% CI 1.00-4.10, p?<?0.05) and higher UICC tumor stages (HR?=?2.21, 95% CI 1.14-4.26, p <?0.05) were associated with impaired survival rates. Split-course treatments (HR?=?2.05, 95% CI 1.07-3.94, p?<?0.05), non-completion of radiotherapy (HR?=?7.17, 95% CI 3.88-13.26, p?<?0.001) and palliative treatments (HR?=?2.84, 95% CI 1.68-4.81, p?<?0.05) were found to result in significantly reduced OS. In the multivariate analysis, split-course concepts (HR?=?2.21, 95% CI 1.10-4.37, p <?0.05) and palliative treatments (HR?=?3.19, 95% CI 1.77-5.75, p <?0.001) significantly deteriorated outcomes, while impaired ECOG status (HR?=?1.49, 95% CI 0.91-2.43, p =?0.11) did not. The vast majority of patients reported either no (n?=?40; 33.6%) or grade 1-2 acute toxicities (n?=?66; 55.5%), and only very few higher-grade toxicities were observed in our study. CONCLUSION:Radiotherapy for nonagenarian patients is generally feasible and associated with a low toxicity profile. Given the relatively poor OS rates and the importance of the quality of life for this patient group, individualized treatment regimens including hypofractionation concepts should be considered.

Project description:Biological age is increasingly recognized as being more accurate than chronological age in determining chronic health outcomes. This study assessed whether biological age, assessed on intensive care unit (ICU) admission, can predict hospital mortality. This retrospective cohort study, conducted in a tertiary multidisciplinary ICU in Western Australia, used the Levine PhenoAge model to estimate each patient's biological age (also called PhenoAge). Each patient's PhenoAge was calibrated to generate a regression residual which was equivalent to biological age unexplained by chronological age in the local context. PhenoAgeAccel was a dichotomized measure of the residuals, and its presence suggested that one was biologically older than the corresponding chronological age. Of the 2950 critically ill adult patients analyzed, 291 died (9.9%) before hospital discharge. Both PhenoAge and its residuals (after regressing on chronological age) had a significantly better ability to differentiate between hospital survivors and non-survivors than chronological age (area under the receiver-operating-characteristic curve 0.648 and 0.654 vs. 0.547 respectively). Being phenotypically older than one's chronological age was associated with an increased risk of mortality (PhenoAgeAccel hazard ratio [HR] 1.997, 95% confidence interval [CI] 1.568-2.542; p = 0.001) in a dose-related fashion and did not reach a plateau until at least a 20-year gap. This adverse association remained significant (adjusted HR 1.386, 95% CI 1.077-1.784; p = 0.011) after adjusted for severity of acute illness and comorbidities. PhenoAgeAccel was more prevalent among those with pre-existing chronic cardiovascular disease, end-stage renal failure, cirrhosis, immune disease, diabetes mellitus, or those treated with immunosuppressive therapy. Being phenotypically older than one's chronological age was more common among those with comorbidities, and this was associated with an increased risk of mortality in a dose-related fashion in the critically ill that was not fully explained by comorbidities and severity of acute illness.

Dataset Information

Prediction of chronological and biological age from laboratory data.

Publications

Prediction of chronological and biological age from laboratory data.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets