publications

2025

1. Prediction of brain age using structural magnetic resonance imaging: A comparison of clinical validity of publicly available software packages

   Ruben P. Dörfel, Brice Ozenne, Melanie Ganz, and 2 more authors

   DOI: 10.1101/2025.03.13.25323902

   medRxiv, Mar 2025

   Abs

   Predicting brain age from structural magnetic resonance images is commonly used as a biomarker of biological aging and brain health. Ideally, as a clinically valid biomarker, brain age should indicate the current state of health and also be predictive of future disease onset and detrimental changes in brain biology. In this preregistered study, we evaluated and compared the clinical validity, i.e., diagnostic and prognostic abilities, of six publicly available brain age prediction packages using data from the Alzheimer’s Disease Neuroimaging Initiative (ADNI). Baseline brain age significantly differed between groups consisting of individuals with normal cognitive function, mild cognitive impairment, and Alzheimer\textquoterights Disease for all packages, but with comparable performance to gray matter volume. Further, estimates were not centered around zero for cognitively normal subjects and showed considerable variation between packages. Finally, brain age did not show strong correlation with disease onset, memory decline, nor gray matter atrophy within four years from baseline in individuals without neurodegenerative disease. The substantial bias and variability in cross-sectional predictions, combined with the weak associations between brain age and longitudinal changes in memory performance or grey matter volume, suggested limited validity of current brain age estimates as a biomarker for biological aging.

2. Evaluation of rapamycin as a neuroprotective treatment in Alzheimer’s disease: a six-month single-arm open-label clinical pilot trial

   Jonas E Svensson, Ruben Paul Dörfel, Martin Schain, and 6 more authors

   DOI: 10.64898/2025.12.08.25340853

   medRxiv, Mar 2025

   Abs

   Background: Rapamycin has demonstrated neuroprotective effects in preclinical Alzheimer’s disease (AD) models, yet clinical data remain limited. Here, we conducted a pilot trial to assess the feasibility and safety of rapamycin treatment in early-stage AD using multimodal neuroimaging and cerebrospinal fluid (CSF) biomarkers. Methods: This single-arm, open-label pilot study enrolled 14 participants with early-stage AD who received oral rapamycin 7 mg weekly for 26 weeks. Thirteen participants completed treatment. The primary outcome was change in cerebral glucose metabolism measured by [18F]FDG PET (Ki and SUVR) in pre-specified regions commonly affected in AD: posterior cingulate cortex, precuneus, and temporoparietal cortex. Secondary outcomes included cerebral blood flow (CBF) via ASL MRI, concentration of CSF biomarkers, volumetric MRI, and cognition. Results: Rapamycin was well tolerated, with no serious adverse events. No significant metabolic or perfusion changes were observed in primary regions. Exploratory analyses showed increases in [18F]FDG SUVR in the putamen, insula, and anterior cingulate cortex, and thalamic CBF. Higher rapamycin blood concentrations correlated with increased [18F]FDG SUVR in several brain regions, including the temporoparietal cortex. CSF analysis showed significant increases in total tau, neurofilament light, ABeta40, and a numerical, non-significant increase in Abeta42 of similar effect size, while p-tau remained largely unchanged, resulting in a significantly decreased p-tau/total tau ratio. Conclusions: This pilot study demonstrates feasibility of rapamycin trials in AD. The primary [18F]FDG outcomes showed no decline, in contrast to what would be expected from natural disease progression The observed elevations in CSF biomarkers warrant further investigation.Competing Interest StatementThe authors have declared no competing interest.Clinical TrialNCT06022068Clinical Protocolshttps://link.springer.com/article/10.1186/s12883-024-03596-1Funding StatementThe study was supported by a Longevity Impetus grant from the Norn Group, Åhlen Stiftelsen, Demensfonden, The Swedish Society of Medicine (SLS), Åke Wibergs Stiftelse (M24-0117), Loo and Hans Osterman Stiftelse, Stiftelsen för Ålderssjukdomar Karolinska Institutet, Stiftelsen för Gamla Tj änarinnor, Tore Nilssons Stiftelse för Medicinsk Forskning, Magnus Bergvalls stiftelse, Karolinska Institutet Research Grants, Stiftelsen Stockholms Sjukhem, Region Stockholm (ALF grant), The Swedish Brain Foundation (PS2021-0012 and PD2024-0444-HK-155), and KI CIMED. None of the funding bodies had any role in the design of the study or in writing the manuscript.Author DeclarationsI confirm all relevant ethical guidelines have been followed, and any necessary IRB and/or ethics committee approvals have been obtained.YesThe details of the IRB/oversight body that provided approval or exemption for the research described are given below:The study was approved by the Swedish Medical Products Agency (5.1-2023-8283) and the Swedish Ethical Review Authority (2023-03075-02 and 2023-00611-01). Written informed consent was obtained from all participants and their designated study partners before any study procedures were initiated. The trial was registered at ClinicalTrials.gov (NCT06022068) and EudraCT (2023-000127-36). I confirm that all necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived, and that any patient/participant/sample identifiers included were not known to anyone (e.g., hospital staff, patients or participants themselves) outside the research group so cannot be used to identify individuals.YesI understand that all clinical trials and any other prospective interventional studies must be registered with an ICMJE-approved registry, such as ClinicalTrials.gov. I confirm that any such study reported in the manuscript has been registered and the trial registration ID is provided (note: if posting a prospective study registered retrospectively, please provide a statement in the trial ID field explaining why the study was not registered in advance).Yes I have followed all appropriate research reporting guidelines, such as any relevant EQUATOR Network research reporting checklist(s) and other pertinent material, if applicable.YesAll data produced in the present study are available upon reasonable request to the authors

3. Machine Learning for Longitudinal Brain-Age Prediction

   Manuel Wegmann, Melanie Ganz, Jonas E. Svensson, and 2 more authors

   DOI: 10.64898/2025.12.10.25341964

   medRxiv, Mar 2025

   Abs

   Cross-sectional brain age models have demonstrated high accuracy and reliability for predicting chronological age based on structural brain features derived from single MRI scans. However, these models cannot separate baseline variation from true aging-related changes or noise. Longitudinal models address this limitation by predicting inter-scan intervals from paired MRI scans, controlling for baseline factors through repeated measurements. Using OASIS-3 data, we compare a cross-sectional 3D CNN against three longitudinal architectures for predicting inter-scan intervals: LILAC (Siamese neural network), LILAC+ (enhanced Siamese network with multi-layer perceptron), and AM (variational autoencoder). Longitudinal models substantially outperformed the cross-sectional approach, with LILAC+ achieving best performance (MSE = 1.97 years^2, MAE = 0.99 years, r = 0.86, R^2 = 0.71). Our results suggest that direct modeling of longitudinal change is more effective at capturing individual aging trajectories than deriving intervals from cross-sectional predictions.Competing Interest StatementThe authors have declared no competing interest.Funding StatementThis study did not receive any fundingAuthor DeclarationsI confirm all relevant ethical guidelines have been followed, and any necessary IRB and/or ethics committee approvals have been obtained.YesThe details of the IRB/oversight body that provided approval or exemption for the research described are given below:The study used ONLY openly available human data that were originally located at: https://www.nitrc.org/projects/oasis3/I confirm that all necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived, and that any patient/participant/sample identifiers included were not known to anyone (e.g., hospital staff, patients or participants themselves) outside the research group so cannot be used to identify individuals.YesI understand that all clinical trials and any other prospective interventional studies must be registered with an ICMJE-approved registry, such as ClinicalTrials.gov. I confirm that any such study reported in the manuscript has been registered and the trial registration ID is provided (note: if posting a prospective study registered retrospectively, please provide a statement in the trial ID field explaining why the study was not registered in advance).Yes I have followed all appropriate research reporting guidelines, such as any relevant EQUATOR Network research reporting checklist(s) and other pertinent material, if applicable.YesAll data produced in the present study are available upon reasonable request to the authorshttps://github.com/manuelwegmann/longitudinal_brainage

2024

1. Multimodal Brain Age Prediction Using Machine Learning: Combining Structural MRI and 5-HT2AR PET-derived Features

   Ruben P. Dörfel, Joan M. Arenas-Gomez, Claus Svarer, and 4 more authors

   DOI: 10.1007/s11357-024-01148-6

   GeroScience, Apr 2024

   Abs

   To better assess the pathology of neurodegenerative disorders and the efficacy of neuroprotective interventions, it is necessary to develop biomarkers that can accurately capture age-related biological changes in the human brain. Brain serotonin 2A receptors (5-HT2AR) show a particularly profound age-related decline and are also reduced in neurodegenerative disorders, such as Alzheimer’s disease. This study investigates whether the decline in 5-HT2AR binding, measured in vivo using positron emission tomography (PET), can be used as a biomarker for brain aging. Specifically, we aim to (1) predict brain age using 5-HT2AR binding outcomes, (2) compare 5-HT2AR-based predictions of brain age to predictions based on gray matter (GM) volume, as determined with structural magnetic resonance imaging (MRI), and (3) investigate whether combining 5-HT2AR and GM volume data improves prediction. We used PET and MR images from 209 healthy individuals aged between 18 and 85 years (mean\,= 38, std\,= 18) and estimated 5-HT2AR binding and GM volume for 14 cortical and subcortical regions. Different machine learning algorithms were applied to predict chronological age based on 5-HT2AR binding, GM volume, and the combined measures. The mean absolute error (MAE) and a cross-validation approach were used for evaluation and model comparison. We find that both the cerebral 5-HT2AR binding (mean MAE\,= 6.63 years, std\,= 0.74 years) and GM volume (mean MAE\,= 6.95 years, std\,= 0.83 years) predict chronological age accurately. Combining the two measures improves the prediction further (mean MAE\,= 5.54 years, std\,= 0.68). In conclusion, 5-HT2AR binding measured using PET might be useful for improving the quantification of a biomarker for brain aging.

2023

1. Prediction of Brain Age Using Structural Magnetic Resonance Imaging: A Comparison of Accuracy and Test–Retest Reliability of Publicly Available Software Packages

   Ruben P. Dörfel, Joan M. Arenas-Gomez, Patrick M. Fisher, and 4 more authors

   DOI: 10.1002/hbm.26502

   Human Brain Mapping, Apr 2023

   Abs

   Brain age prediction algorithms using structural magnetic resonance imaging (MRI) aim to assess the biological age of the human brain. The difference between a person’s chronological age and the estimated brain age is thought to reflect deviations from a normal aging trajectory, indicating a slower or accelerated biological aging process. Several pre-trained software packages for predicting brain age are publicly available. In this study, we perform a comparison of such packages with respect to (1) predictive accuracy, (2) test–retest reliability, and (3) the ability to track age progression over time. We evaluated the six brain age prediction packages: brainageR, DeepBrainNet, brainage, ENIGMA, pyment, and mccqrnn. The accuracy and test–retest reliability were assessed on MRI data from 372 healthy people aged between 18.4 and 86.2 years (mean 38.7 \textpm 17.5 years). All packages showed significant correlations between predicted brain age and chronological age (r = 0.66–0.97, p < 0.001), with pyment displaying the strongest correlation. The mean absolute error was between 3.56 (pyment) and 9.54 years (ENIGMA). brainageR, pyment, and mccqrnn were superior in terms of reliability (ICC values between 0.94–0.98), as well as predicting age progression over a longer time span. Of the six packages, pyment and brainageR consistently showed the highest accuracy and test–retest reliability.