Access to this full-text is provided by Springer Nature.
Content available from Nature Medicine
This content is subject to copyright. Terms and conditions apply.
Articles
https://doi.org/10.1038/s41591-022-01980-3
1Center for Digital Health, Berlin Institute of Health at Charité – Universitätsmedizin Berlin, Berlin, Germany. 2Department of Cardiology, Campus Benjamin
Franklin, Charité – Universitätsmedizin Berlin and Berlin Institute of Health, Berlin, Germany. 3Computational Medicine, Berlin Institute of Health at
Charité – Universitätsmedizin Berlin, Berlin, Germany. 4MRC Epidemiology Unit, Institute of Metabolic Science, University of Cambridge, Cambridge,
UK. 5Molecular Epidemiology, LUMC, Leiden, the Netherlands. 6Leiden Computational Biology Center, LUMC, Leiden, The Netherlands. 7Department of
Epidemiology, Erasmus MC University Medical Center, Rotterdam, the Netherlands. 8Molecular Epidemiology, Department of Biomedical Data Sciences,
Leiden University Medical Center, Leiden, the Netherlands. 9Institute of Cardiovascular Sciences, University College London, London, UK. 10Department of
Endocrinology & Metabolism, Charité – Universitätsmedizin Berlin and Berlin Institute of Health, Berlin, Germany. 11Center for Cardiovascular Research,
Charité – Universitätsmedizin Berlin and Berlin Institute of Health, Berlin, Germany. 12Department of Neurology, Humboldt-Universität zu Berlin and
Berlin Institute of Health, Charité-Universitätsmedizin Berlin, Berlin, Germany. 13School of Mind and Brain, Humboldt-Universität zu Berlin, Berlin,
Germany. 14Delft Bioinformatics Lab, TU Delft, Delft, the Netherlands. 15Department of Internal Medicine, Division of Gerontology and Geriatrics, Leiden
University Medical Center, Leiden, the Netherlands. 16Department of Cardiology, Leiden University Medical Center, Leiden, the Netherlands. 17Institute
of Cardiovascular and Medical Sciences, Cardiovascular Research Centre, University of Glasgow, Glasgow, UK. 18Netherlands Heart Institute, Utrecht,
the Netherlands. 19Department of Surgery, Erasmus MC University Medical Center, Rotterdam, the Netherlands. 20Max Planck Institute for the Biology
of Ageing, Cologne, Germany. 21Department of Epidemiology and Public Health, University College London, London, UK. 22Clinicum, Faculty of Medicine,
University of Helsinki, Helsinki, Finland. 23Health Data Science Unit, Heidelberg University Hospital and BioQuant, Heidelberg, Germany. 24These authors
contributed equally: Thore Buergel, Jakob Steinfeldt. 25These authors jointly supervised this work: John Deanfield, Roland Eils, Ulf Landmesser.
✉e-mail: roland.eils@bih-charite.de
Risk stratification is central to disease prevention1,2. Over the
past decade, increasingly complex information on an indi-
vidual’s phenotype has become available beyond conventional
demographic and laboratory information. While blood metabolites
such as cholesterols are established clinical predictors for cardio-
vascular disease risk3, many more have been linked to common
disease phenotypes4–8. In recent years, studies have moved beyond
associations of individual markers by linking metabolomic profiles
to aging9, disease onset10 and mortality11, appreciating the human
blood metabolome as a direct reflection of the physiological state.
Proton nuclear magnetic resonance (1H-NMR) spectroscopy
enables a standardized assessment of a multitude of small circulat-
ing molecules in the blood simultaneously. NMR differs from other
techniques in metabolomics, such as mass spectrometry, by its vir-
tual absence of batch effects, minimal requirements of expensive
reagents and high throughput at comparatively low cost12. In the
current assay >150 original markers are quantified, including amino
and fatty acids and metabolites related to carbohydrate metabolism
and fluid balance, partly overlapping with conventional clinical pre-
dictors including glucose, albumin and creatinine13–15. Further, the
Metabolomic profiles predict individual
multidisease outcomes
Thore Buergel 1,24, Jakob Steinfeldt2,24, Greg Ruyoga1, Maik Pietzner3,4, Daniele Bizzarri 5,6,
Dina Vojinovic7,8, Julius Upmeier zu Belzen 1, Lukas Loock1, Paul Kittner1, Lara Christmann1,
Noah Hollmann 1, Henrik Strangalies1, Jana M. Braunger1, Benjamin Wild 1, Scott T. Chiesa 9,
Joachim Spranger 10,11, Fabian Klostermann12,13, Erik B. van den Akker 5,6,14, Stella Trompet 15,16,
Simon P. Mooijaart15, Naveed Sattar 17, J. Wouter Jukema 16,18, Birgit Lavrijssen7,19, Maryam Kavousi7,
Mohsen Ghanbari 7, Mohammad A. Ikram 7, Eline Slagboom5,20, Mika Kivimaki 21,22,
Claudia Langenberg3,4, John Deanfield9,25, Roland Eils 1,23,25 ✉ and Ulf Landmesser 2,25
Risk stratification is critical for the early identification of high-risk individuals and disease prevention. Here we explored the
potential of nuclear magnetic resonance (NMR) spectroscopy-derived metabolomic profiles to inform on multidisease risk
beyond conventional clinical predictors for the onset of 24 common conditions, including metabolic, vascular, respiratory, mus-
culoskeletal and neurological diseases and cancers. Specifically, we trained a neural network to learn disease-specific metabo-
lomic states from 168 circulating metabolic markers measured in 117,981 participants with ~1.4 million person-years of follow-up
from the UK Biobank and validated the model in four independent cohorts. We found metabolomic states to be associated with
incident event rates in all the investigated conditions, except breast cancer. For 10-year outcome prediction for 15 endpoints,
with and without established metabolic contribution, a combination of age and sex and the metabolomic state equaled or out-
performed established predictors. Moreover, metabolomic state added predictive information over comprehensive clinical vari-
ables for eight common diseases, including type 2 diabetes, dementia and heart failure. Decision curve analyses showed that
predictive improvements translated into clinical utility for a wide range of potential decision thresholds. Taken together, our
study demonstrates both the potential and limitations of NMR-derived metabolomic profiles as a multidisease assay to inform
on the risk of many common diseases simultaneously.
NATURE MEDICINE | VOL 28 | NOVEMBER 2022 | 2309–2320 | www.nature.com/naturemedicine 2309
Content courtesy of Springer Nature, terms of use apply. Rights reserved
Articles Nature MediciNe
assay has a high resolution of lipoprotein particles, measuring their
components, sizes and concentrations13,14. This high-throughput
NMR metabolomics platform has been explored in multiple studies
investigating all-cause mortality11,16, cardiovas cular disease13,17, type 2
diabetes (T2D)18,19, Alzheimer’s disease8 and COVID-19 (ref. 20).
Importantly, recent work has indicated a broad metabolic basis
across diseases, suggesting a shared etiology21. This systemic infor-
mation contained in metabolomic profiles has been insufficiently
considered in the risk prediction of common diseases.
Here we exploited the potential of NMR-based blood profiling as
a single-domain assay to simultaneously predict multidisease onset.
We developed, trained and validated a deep residual multitask
neural network to simultaneously learn disease-specific metabo-
lomic states for 24 conditions, including common metabolic, vas-
cular, respiratory, musculoskeletal and neurological disorders and
cancers (Fig. 1). The scalar metabolomic states, contained in a
24-dimensional vector, were derived from 168 circulating metabo-
lomic markers measured in ~120.000 individuals in the UK Biobank
population cohort22. We extensively investigated the learned metab-
olomic states by integrating them in Cox proportional hazard (CPH)
models23, modeling the risk for individual endpoints and demon-
strating that information gained through NMR metabolomic profil-
ing is additive to known clinical predictors. Moreover, we externally
validated the metabolomic states in four independent cohorts, the
Whitehall II cohort24, the Rotterdam Study25, the Leiden Longevity
Study26 and the PROspective Study of Pravastatin in the Elderly at
Risk27 (Fig. 1c), and investigated their clinical utility.
Results
Study population and the metabolomic state model. Based on the
UK Biobank cohort22,28, we derived an integrated metabolomic state
capturing information on incident disease risk in a general popula-
tion sample (Fig. 1a,b). We extracted clinical predictors and disease
endpoints for 117,981 individuals with serum NMR profiling at the
time of cohort recruitment (Methods and Supplementary Tables
1–3). The study population had a median age of 58 years (interquar-
tile range (IQR) 50, 63), of whom 54.2% were female, 11% current
smokers and 5.2% diagnosed with T2D (Table 1). Median body mass
index (BMI) was 26.8 (IQR 24.2, 29.9), systolic blood pressure was
136 mmHg (IQR 124, 149), total cholesterol was 5.65 mmol l–1 (IQR
4.90, 6.42) and glucose was 4.93 mmol l–1 (IQR 4.60, 5.32). Median
follow-up was 12.2 years with ~1,435,340 overall person-years. To
maximize the generalizability and transferability of our results, we
partitioned the data spatially by the 22 recruitment centers. For each
center, all individuals from a single center were retained for testing of
models that were trained on individuals pooled from the 21 remain-
ing recruitment centers and then randomly split into training and
validation sets to develop the models. After model selection on the
validation datasets and obtaining the selected models’ final predic-
tions on the individual test sets, test set predictions were aggregated
for downstream analysis (Fig. 1b).
We externally validated disease-specific metabolomic states in
four independent cohorts analyzed with the same 1H-NMR metab-
olomics assay, the Whitehall II cohort24, and three independent
cohorts of the BBMRI-NL consortium (Fig. 1c). The Whitehall II
cohort24 is an ongoing prospective cohort study, including metab-
olomics for 6,197 participants aged 44–69 years. The Rotterdam
Study is a prospective, population-based cohort study among indi-
viduals living in the Ommoord district in the city of Rotterdam (the
Netherlands)25, offering metabolomics for 2,949 participants with a
median age of 74 years (IQR 70–79). The Leiden Longevity PAROFF
Study (LLS)26 comprises offspring and spouses of long-lived indi-
viduals, with metabolomics available for 1,655 individuals with a
mean age of 59 years (IQR 54–63). Finally, the PROspective Study
of Pravastatin in the Elderly at Risk (PROSPER) is a clinical trial
investigating pravastatin effects27, of which 960 samples with a
median age of 76 years (IQR 73–78) are included in the BBMRI-NL
platform. Detailed characteristics of the four replication cohorts are
presented in Supplementary Data and Supplementary Table 4.
The metabolomic state model is a multitask residual neural net-
work trained on the entire set of 168 original metabolomic mark-
ers to model the integrative metabolomic state for all 24 endpoints
simultaneously (Fig. 1b, Extended Data Fig. 1 and Metabolomic
state model). This allowed us to leverage the shared metabolite
profiles while retaining flexibility in fitting endpoint-specific varia-
tions, outperforming endpoint-specific linear models and linear
models on principal components (Extended Data Fig. 2).
To test whether multidisease states could be equally informa-
tive from readily accessible information from study participants
at baseline, we investigated three different scenarios with increas-
ingly comprehensive predictor sets. First, we considered age and
sex only, both highly predictive for common diseases and avail-
able at no cost. Second, we investigated cardiovascular predictors
from well-validated primary prevention scores, the American Heart
Association (ASCVD)3, which are easily accessible at minimal cost
and are predictive beyond cardiovascular disease, including neu-
rological and neoplastic conditions29–31. Third, we extended these
predictors with a comprehensive set of clinical predictors beyond
what is typically available in primary care. These included >30 pre-
dictors with information on lifestyle factors, physical measurements
and laboratory values, as well as further validated disease-specific
predictors from FINDRISC32 (T2D) and CAIDE33 (dementia) scores
(Fig. 1d and Supplementary Table 2).
Metabolomic state stratifies the risk of disease onset. A critical
component of prevention is identification of individuals at high risk
of developing a disease, often at an early subclinical stage. To inves-
tigate whether the NMR-derived metabolomic state informs disease
risk, we assessed the link with incident event rates in the observa-
tion period (Fig. 2a). To allow comparison between the endpoints
despite the large differences in event rates (Supplementary Table 7;
for example, Parkinson’s disease, 0.6%; major adverse cardiac event
(MACE), 8.7%), we also calculated the observed event rate ratio
between individuals in the top and bottom 10% of metabolomic
states (Fig. 2 and Supplementary Table 7) with 95% confidence
intervals (CIs).
We observed increasing event rates over metabolomic state per-
centiles for all 24 investigated diseases, except breast cancer. For 15 of
the 24 diseases, the top 10% of the metabolomic state corresponded
to a rate more than fivefold higher compared with the bottom 10%.
For conditions such as T2D (top 10%, 21.87%; bottom 10%, 0.36%;
odds ratio (OR) 61.45, 95% CI 47.00, 86.12), abdominal aortic aneu-
rysm (AAA) (top 10%, 2.46%; bottom 10%, 0.18%; OR 14.1, 95%
CI 9.93, 24.45) and heart failure (top 10%, 10.80%; bottom 10%,
0.96%; OR 11.27, 95% CI 9.43, 13.50) the ratio was >10. Ratios for
most other diseases were lower—for example, cerebral stroke 9.66
(95% CI 7.64, 12.14), MACE 9.25 (95% CI 8.12, 10.53), atrial fibril-
lation 8.13 (95% CI 6.95, 9.37), all-cause dementia 6.39 (95% CI
5.40, 8.09) or chronic obstructive pulmonary disease (COPD) 4.98
(95% CI 4.37, 5.62). In contrast, we observed much smaller ratios
for some diseases—for example, glaucoma (top 10%, 3.47%; bottom
10%, 1.57%; OR 2.19, 95% CI 1.91, 2.62) or asthma (top 10%, 5.52%;
bottom 10%, 2.48%; OR 2.22, 95% CI 2.01, 2.57), thus suggesting
less information contained in the respective metabolomic states. In
summary, the disease-specific metabolomic state stratified risk tra-
jectories for all investigated endpoints except breast cancer (Fig. 2b),
separating the rates of cumulative events most notably for T2D,
renal disease and heart failure but also, to a much lesser extent, for
glaucoma or asthma.
Information is shared with clinical predictors. Many clinical pre-
dictors are readily available in primary care and commonly used to
NATURE MEDICINE | VOL 28 | NOVEMBER 2022 | 2309–2320 | www.nature.com/naturemedicine
2310
Content courtesy of Springer Nature, terms of use apply. Rights reserved
Articles
Nature MediciNe
117,981 Individuals
with NMR metabolomics
at baseline
Clinical
predictors
NMR
metabolomics:
168 circulating
metabolites
Metabolomic state
per endpoint
Machine learning
model
Survival analysis per endpoint
Metabolomic state
Clinical predictor set X
+ metabolomic state
Clinical predictor set X
CPH(X, )
CPH(X)
CPH( )
22
Recruitment
centers
From
All individuals
from a center are
assigned to test
All other individuals
are randomly assigned
to training or validation sets
Test Training Validation
80% 20%
Benchmarking
per endpoint
Test set
Training and
validation sets
Model development for each of the 22 partitions
22 Training,
validation
and test splits
Aggregated
predictions of
22 test sets
Dataset preparation
Metabolomic state model
Combined
downstream
analysis
a
b
d
Age
Sex
Smoking
Alcohol intake
Physical activity
Education years
Daily healthy food
FH T2D
T2D
BMI
Waist/hip ratio
Waist circumference
Weight
Standing height
Systolic blood pressure
Total cholesterol
LDL cholesterol
HDL cholesterol
Triglycerides
Glucose
HbA1c
Creatinine
Cystatin C
Urea
Urate
Aspartate aminotransferase
Alkaline phosphatase
Albumin
C-reactive protein
Erythrocyte count
Leukocyte count
Platelet count
Hemoglobin
Hematocrit
Mean corpuscular Hb
Mean corpuscular volume
Mean corpuscular Hb concentration
Antihypertensives
Age+Sex
ASCVD
PANEL
FINDRISC
CAIDE
Alanine aminotransferase
Demography
and lifestyle
Familial
history
Diagnoses
Physical measurements Lipids
Glucose metabolism
Kidney Liver
Inflammation
Blood counts
Medication
NMR
metabolomics
Trained metabolomic state model
Application of
metabolomic state model Clinical
predictors
Benchmarking
per endpoint
Whitehall II
cohort
(n = 6,117)
cExternal validation in four independent cohorts
Rotterdam
study
(n = 2,949)
PROSPER
cohort
(n = 960)
Leiden longevity
study
(n = 1,655)
Survival analysis per endpoint
CPH(X, )
CPH(X)
CPH( )
For each
center
Fig. 1 | Study overview. a, To learn metabolomic states from circulating blood metabolites, the eligible UK Biobank population (with NMR blood
metabolomics and valid consent) was split into training, validation and test sets with 22-fold nested cross-validation based on the assigned UK Biobank
assessment center. b, For each of the 22 partitions, the metabolomic state model was trained on the 168 metabolomic markers to predict metabolomic risk
against 24 common disease endpoints. Subsequently, for each endpoint, CPH models were developed on the metabolomic state in combination with sets
of commonly available clinical predictors to model disease risk. Predictions of the CPH model on the test set were aggregated for downstream analysis.
c, The metabolomic state model was externally validated in four independent cohorts—the Whitehall II cohort and three from the BBMRI-NL consortium:
the Rotterdam Study, the Leiden Longevity Study and the PROSPER cohort. d, In this study we consider clinical predictors from scores commonly applied
in primary prevention. We additionally integrate variables into a comprehensive predictor set (PANEL) to investigate overlapping information with the
metabolomic state. FH, family history.
NATURE MEDICINE | VOL 28 | NOVEMBER 2022 | 2309–2320 | www.nature.com/naturemedicine 2311
Content courtesy of Springer Nature, terms of use apply. Rights reserved
Articles Nature MediciNe
stratify the risk of common diseases such as cardiovascular disease3,
kidney disease34 or diabetes32. While more complex risk scores have
been proposed35, the trade-off between the added predictive infor-
mation and resources in time and cost required to collect the new
data has limited clinical adoption36. We therefore investigated the
predictive information of the relatively affordable and standardized
NMR metabolomics assay against common clinical variables in the
UK Biobank and in four independent validation cohorts.
First, we modeled disease risk for each endpoint in the UK
Biobank using CPH models for three clinical predictor sets with
increasing complexity: Age+Sex, highly predictive and available
ahead of any test; ASCVD, a set of readily available cardiovascular
predictors; and PANEL, a comprehensive selection of clinical pre-
dictors including in-depth blood measurements (Fig. 1d) exceeding
those typically available in primary care. For all sets, the perfor-
mance of CPH models was benchmarked against those based on
the sets’ combinations with the metabolomic state. As quantified
by Harrell’s C-index, the discriminative performances of all mod-
els at 10 years after baseline are shown in Fig. 3a. Subsequently,
to validate metabolomic states, we applied the trained metabolo-
mic state model to the external validation cohorts and replicated
the CPH models with and without metabolomic state addition for
Table 1 | The study population
Characteristic Male, n = 54,078aFemale, n = 63,903aOverall, n = 117,981a
Age at recruitment 58 (50, 64) 57 (50, 63) 58 (50, 63)
Education years 15.00 (11.00, 15.00) 13.00 (11.00, 15.00) 13.00 (11.00, 15.00)
Current smoker 6,724 (12%) 5,747 (9.0%) 12,471 (11%)
Daily alcohol intake 13,651 (25%) 10,191 (16%) 23,842 (20%)
Daily moderate to vigorous physical
activity 50 (15, 105) 45 (10, 90) 45 (10, 90)
Daily healthy food 52,974 (98%) 63,290 (99%) 116,264 (99%)
Family history of diabetes 8,827 (16%) 11,266 (18%) 20,093 (17%)
T2D 3,882 (7.2%) 2,295 (3.6%) 6,177 (5.2%)
Weight (kg) 84 (76, 94) 69 (62, 79) 76 (66, 88)
Standing height (cm) 176 (171, 180) 162 (158, 167) 168 (162, 175)
BMI 27.3 (25.0, 30.1) 26.1 (23.5, 29.7) 26.8 (24.2, 29.9)
Waist/hip ratio 0.93 (0.89, 0.98) 0.81 (0.77, 0.86) 0.87 (0.80, 0.94)
Waist circumference (cm) 96 (89, 103) 83 (76, 92) 90 (80, 99)
Systolic blood pressure (mmHg) 139 (128, 152) 133 (121, 147) 136 (124, 149)
Total cholesterol (mmol l–1) 5.45 (4.70, 6.21) 5.80 (5.07, 6.58) 5.65 (4.90, 6.42)
LDL cholesterol (mmol l–1) 3.46 (2.87, 4.05) 3.56 (3.00, 4.17) 3.52 (2.94, 4.12)
HDL cholesterol (mmol l–1) 1.24 (1.06, 1.45) 1.55 (1.32, 1.82) 1.40 (1.17, 1.67)
Triglycerides (mmol l–1) 1.69 (1.18, 2.44) 1.33 (0.96, 1.89) 1.48 (1.04, 2.14)
Glucose (mmol l–1) 4.96 (4.61, 5.37) 4.91 (4.59, 5.28) 4.93 (4.60, 5.32)
Glycated hemoglobin (%) 35.3 (32.8, 38.1) 35.2 (32.7, 37.7) 35.2 (32.8, 37.9)
Creatinine (umol l–1) 80 (72, 88) 63 (57, 70) 70 (61, 81)
Cystatin C (mg l–1) 0.92 (0.84, 1.01) 0.86 (0.78, 0.95) 0.88 (0.80, 0.98)
Urea (mmol l–1) 5.45 (4.68, 6.33) 5.10 (4.33, 5.95) 5.26 (4.49, 6.13)
Urate (umol l–1) 350 (305, 399) 264 (225, 309) 303 (250, 361)
Aspartate aminotransferase (U l–1) 26 (23, 31) 23 (20, 27) 24 (21, 29)
Alanine aminotransferase (U l–1) 24 (18, 32) 18 (14, 23) 20 (15, 27)
Alkaline phosphatase (U l–1) 79 (67, 93) 82 (67, 98) 80 (67, 96)
Albumin (g l–1) 45.52 (43.80, 47.24) 44.91 (43.21, 46.63) 45.20 (43.47, 46.93)
C-reactive protein (mg l–1) 1.29 (0.67, 2.55) 1.38 (0.65, 2.95) 1.33 (0.66, 2.76)
Erythrocytes (1012 cells l–1) 4.74 (4.51, 4.98) 4.32 (4.10, 4.54) 4.50 (4.23, 4.79)
Leukocytes (109 cells l–1) 6.68 (5.66, 7.89) 6.61 (5.60, 7.81) 6.64 (5.62, 7.85)
Platelets (109 cells l–1) 234 (202, 269) 261 (226, 301) 248 (214, 287)
Hemoglobin (g dl–1) 15.00 (14.37, 15.64) 13.50 (12.90, 14.10) 14.15 (13.31, 15.02)
Hematocrit (%) 43.3 (41.4, 45.2) 39.2 (37.5, 41.0) 41.0 (38.7, 43.5)
Mean corpuscular volume (fl) 91.4 (88.8, 94.1) 91.1 (88.4, 93.7) 91.2 (88.6, 93.9)
Mean corpuscular hemoglobin (pg) 31.69 (30.70, 32.70) 31.37 (30.33, 32.37) 31.50 (30.50, 32.50)
Mean corpuscular hemoglobin (g dl–1) 34.60 (34.00, 35.22) 34.36 (33.80, 35.00) 34.48 (33.90, 35.10)
Antihypertensives 1,090 (2.0%) 680 (1.1%) 1,770 (1.5%)
aMedian (IQR); n (%)
NATURE MEDICINE | VOL 28 | NOVEMBER 2022 | 2309–2320 | www.nature.com/naturemedicine
2312
Content courtesy of Springer Nature, terms of use apply. Rights reserved
Articles
Nature MediciNe
the Age+Sex predictor set for all endpoints available. The results
of the external validation are shown in Extended Data Fig. 3. We
noted the discriminative performance of the metabolomic state
to be highly disease dependent: while the metabolomic state
contained significantly less predictive information than clinical
predictors for cataract, glaucoma and skin, colon, rectal and pros-
tate cancers, this was not the case for renal disease, liver disease and
T2D. Here, the metabolomic state contained a greater predictive
value than Age+Sex and even ASCVD. Generally, we observed an
increase in discriminative performance with the addition of more
Lung cancer Skin cancer Colon cancer Rectal cancer Prostate cancer Breast cancer
COPD Asthma Parkinson’s Cataract Glaucoma Fracture
T2D Liver disease Renal disease PAD Venous thrombosis AAA
25 50 75 100 25 50 75 100 25 50 75 100 25 50 75 100 25 50 75 100 25 50 75 100
25 50 75 100 25 50 75 100 25 50 75 100 25 50 75 100 25 50 75 100 25 50 75 100
25 50 75 100 25 50 75 100 25 50 75 100 25 50 75 100 25 50 75 100 25 50 75 100
0
5
10
15
20
0
1
2
3
4
0
4
8
12
16
0
1
2
3
0
5
10
15
0
1
2
3
4
5
0
1
2
3
4
5
0
1
2
3
4
5
0
2.5
5.0
7.5
0
5
10
15
0
5
10
15
20
0
0.4
0.8
1.2
1.6
0
2
4
6
0
10
20
30
40
50
0
0.5
1.0
1.5
0
1
2
0
5
10
15
20
0
5
10
15
20
0
2
4
6
8
0
2
4
6
8
MACE CHD Cerebral stroke Dementia Heart failure Atrial fibrillation
25 50 75 100 25 50 75 100 25 50 75 100 25 50 75 100 25 50 75 100 25 50 75 100
0
5
10
15
20
25
0
10
20
30
40
50
0
5
10
15
20
25
0
1
2
3
4
5
Metabolomics state percentile (%)
Observed event rate (%)Cumulative events (%)
Lung cancer Skin cancer Colon cancer Rectal cancer Prostate cancer Breast cancer
COPD Asthma Parkinson’s Cataract Glaucoma Fracture
T2D Liver disease Renal disease PAD Venous thrombosis AAA
5 10 5 10 5 10 5 10 5 10 5 10
5 10 5 10 5 10 5 10 5 10 5 10
5 10 5 10 5 10 5 10 5 10 5 10
0
4
8
12
16
0
1
2
3
4
0
5
10
0
1
2
0
2.5
5.0
7.5
10.0
12.5
0
1
2
3
0
1
2
3
4
0
1
2
3
4
0
2
4
6
8
0
3
6
9
0
5
10
15
20
0
0.5
1.0
0
2
4
6
0
10
20
30
0
1
2
3
0
0.5
1.0
1.5
2.0
0
5
10
15
20
0
4
8
12
0
2
4
6
0
2
4
6
MACE CHD Cerebral stroke Dementia Heart failure Atrial fibrillation
5 10 5 10 5 10 5 10 5 10 5 10
0
5
10
15
20
0
5
10
15
20
0
5
10
15
20
0
1
2
3
4
Time (years)
a
b
Top 10%Metabolomic state quantile Bottom 10%Median 10%
Fig. 2 | Metabolomic state is associated with ORs and stratifies survival. a, Observed event frequency for incident disease plotted against metabolomic
state percentiles over the entire study population for all 24 endpoints. b, Cumulative event rates over the observation time for all assessed endpoints,
stratified by metabolomic state quantiles (light blue, bottom 10%; blue, median 10%; dark blue, top 10%), with 95% CIs indicated. PAD, peripheral
artery disease.
NATURE MEDICINE | VOL 28 | NOVEMBER 2022 | 2309–2320 | www.nature.com/naturemedicine 2313
Content courtesy of Springer Nature, terms of use apply. Rights reserved
Articles Nature MediciNe
Liver disease
Renal disease
Heart failure
Lung cancer
Venous thrombosis
Breast cancer
Cerebral stroke
Atrial fibrillation
Rectal cancer
Skin cancer
Parkinson’s
Prostate cancer
–0.10 –0.05 0 0.05 0.10
ASCVD
–0.04 –0.02 0 0.02 0.04
PANEL
bd
0.650
0.675
0.700
0.725
0.750
0.675
0.700
0.725
0.750
Absolute C–index
aMACE CHD
0.66
0.68
0.70
Cerebral stroke Dementia
0.62
0.64
0.66
0.68
0.70
0.69
0.72
0.75
0.78
Heart failure Atrial fibrillation
0.66
0.69
0.72
0.75
0.78
T2D Liver disease Renal disease PAD Venous thrombosis AAA
0.70
0.75
0.80
0.60
0.63
0.66
0.69
0.64
0.66
0.68
0.70
0.72
0.62
0.64
0.66
0.68
0.70
0.60
0.65
0.70
0.6
0.7
0.8
COPD Asthma Parkinson’s Cataract Glaucoma Fracture
0.57
0.58
0.59
0.60
0.61
0.550
0.575
0.600
0.625
0.650
0.675
0.625
0.650
0.675
0.700
0.725
0.750
0.60
0.65
0.70
0.75
0.525
0.550
0.575
0.600
0.625
0.60
0.62
0.64
0.66
0.68
Lung cancer Skin cancer Colon cancer Rectal cancer Prostate cancer Breast cancer
MET
Age+Sex
AgeSex + MET
ASCVD
ASCVD + MET
PANEL
PANEL + MET
0.51
0.53
0.55
0.57
0.55
0.60
0.65
0.70
0.60
0.64
0.68
0.60
0.64
0.68
0.72
0.60
0.65
0.70
0.75
0.80
MET
Age+Sex
AgeSex + MET
ASCVD
ASCVD + MET
PANEL
PANEL + MET
MET
Age+Sex
AgeSex + MET
ASCVD
ASCVD + MET
PANEL
PANEL + MET
MET
Age+Sex
AgeSex + MET
ASCVD
ASCVD + MET
PANEL
PANEL + MET
MET
Age+Sex
AgeSex + MET
ASCVD
ASCVD + MET
PANEL
PANEL + MET
MET
Age+Sex
AgeSex + MET
ASCVD
ASCVD + MET
PANEL
PANEL + MET
T2D
Asthma
COPD
CHD
PAD
Fracture
Dementia
MACE
AAA
Colon cancer
Glaucoma
Cataract
–0.2 –0.1 0 0.1 0.2
∆ C-index ∆ C-index ∆ C-index
Age+Sex c
Fig. 3 | Predictive value of the metabolomic state is endpoint dependent. a, Comparison of discriminative performance of CPH models trained on
the metabolomic state only (MET), the three clinical predictor sets (Age+Sex, ASCVD and PANEL) and the sets’ combinations with the metabolomic
state. Horizontal dashed lines indicate the median performance of the three clinical predictor sets. b, Differences in discriminative performance
between the Age+Sex baseline (dashed line), metabolomic state only (blue) and the combination of Age+Sex and metabolomic state (green).
c, Differences in discriminative performance between ASCVD predictors (dashed line), the combination of Age+Sex and the metabolomic state
(green) and the combination of metabolomic state and ASCVD predictors (red). d, Difference in discriminative performance between comprehensive
PANEL predictors (dashed line), ASCVD + MET (red) and PANEL + MET (black). a–d, Statistical measures were derived from n = 117.981 individuals;
those with previous events were excluded (Supplementary Table 1). Data are presented as median (center of error bar) and 95% CI (line of error
bar) determined by bootstrapping of with 1,000 iterations. b–d, The x-axis range differs across panels; vertical grid lines indicate differences of 0.02
C-index.
NATURE MEDICINE | VOL 28 | NOVEMBER 2022 | 2309–2320 | www.nature.com/naturemedicine
2314
Content courtesy of Springer Nature, terms of use apply. Rights reserved
Articles
Nature MediciNe
comprehensive clinical predictors across all endpoints, and perfor-
mances were stable over different age groups, biological sexes and
ethnic backgrounds (Extended Data Fig. 4).
To better assess the predictive value of the metabolomic state
(MET) in comparison with clinical variables, we calculated C-index
deltas (Fig. 3b–d). We noted that CPH models fit solely on the
metabolomic state performed competitively or better than Age+Sex
for ten of the 24 endpoints, including T2D and COPD, but also for
heart failure, liver disease and renal disease (Fig. 3b). The com-
petitive performance compared with Age+Sex was replicated in
the validation cohorts for T2D, COPD, heart failure, coronary heart
disease (CHD) and all-cause dementia (Extended Data Fig. 3 and
Supplementary Table 10).
Interestingly, CPH models fit on the combination of the metab-
olomic state with Age+Sex (Age+Sex + MET) performed com-
parably to, or better than, the ASCVD predictors for 15 of the
24 endpoints, including T2D, liver disease, renal disease, heart
failure, venous thrombosis and dementia (Fig. 3c). While the com-
prehensive PANEL score generally contained the most predictive
information, surprisingly we observed only modest gains over the
combination of ASCVD and the metabolomic state, and Age+Sex
and the metabolomic state (Fig. 3d). Applying the complex metabo-
lomic state model architecture to the predictors of the PANEL, we
did not observe systematic performance improvements (Extended
Data Fig. 5).
Discriminative improvements over clinical predictors. In addi-
tion to investigating the shared information, we were interested in
quantifying the additive predictive value of metabolomic state over
readily available clinical variables. To understand how the infor-
mation is distributed over the PANEL predictors, we first assessed
the aggregated coefficients of the CPH model and found that basic
demographic information, medical history and physical measure-
ments provided the most predictive information over all endpoints
(Supplementary Table 5). In addition, apart from shared measures
(for example, glucose, albumin or creatinine), lipids and creatinine/
cystatin c, we did not observe strong correlations (|r| > 0.5) between
the PANEL predictors and NMR metabolites (Supplementary Table
6). Therefore, we continued assessment of performance differ-
ences between the CPH models’ fit on clinical predictors and those
with the added metabolomic state by calculating differences in the
C-index (Supplementary Table 9).
In the UK Biobank, the metabolomic state significantly added
predictive information over age and sex for 18 of the 24 endpoints;
in contrast, endpoints with a comparably low predictive value of the
metabolomic state, such as Parkinson’s disease, skin cancer, colon
cancer, rectal cancer, glaucoma and cataract, did not benefit from
the addition of the metabolomic state. Results from four exter-
nal cohorts independently confirmed significant discriminative
improvements over Age+Sex for CHD, heart failure, atrial fibrilla-
tion, T2D and COPD (for detailed results and event counts for the
independent cohorts, see Extended Data Fig. 3a and Supplementary
Table 10).
Beyond basic demographic predictors, addition of the metabolo-
mic state to cardiovascular predictors further significantly improved
discriminative performance for 15 of the 24 endpoints (Fig. 3c).
Even when added to the comprehensive PANEL set, the metabo-
lomic state provided significant additional discriminatory value
for eight of the 24 endpoints (Fig. 3d) as quantified by C-index,
including T2D (0.009, 95% CI 0.007, 0.012), dementia (0.005, 95%
CI 0, 0.009), heart failure (0.005, 95% CI 0.003, 0.007), COPD
(0.005, 95% CI 0.003, 0.006), renal disease (0.004, 95% CI 0.002,
0.005), CHD (0.003, 95% CI 0.001, 0.004) and MACE (0.002, 95%
CI 0, 0.004).
We fur ther sought to understand the potential of the metabolomic
state in regard to individual risk under consideration of established
clinical predictors. Therefore, we examined the partial effects and
hazard ratios (HRs, per s.d. metabolomic state, with 95% CI) of
the CPH models trained on the combinations of the metabolomic
state and predictor sets Age+Sex, ASCVD and PANEL (Extended
Data Fig. 6a) for those 18 endpoints with discrimination benefits
over the Age+Sex set. We observed a notable separation between
the top, median and bottom 10% of the metabolomic state in 14
of the 18 endpoints when adjusted for more comprehensive clini-
cal predictors (for HRs, see Extended Data Fig. 6b). A change of
1 s.d. in the metabolomic state for T2D resulted in substantially
adjusted HRs (HRAge+Sex 3.83 (95% CI 3.71–4.01), HRPANEL 2.5
(95% CI 2.34–2.67)), which were replicated with adjustment for
Age+Sex in the independent cohorts (Extended Data Fig. 3b).
Other investigated endpoints, such as all-cause dementia (HRAge+Sex
1.56 (95% CI 1.54–1.72), HRPANEL 1.46 (95% CI 1.43–1.47)), heart
failure (HRAge+Sex 1.8 (95% CI 1.74–1.86), HRPANEL 1.45 (95% CI
1.38–1.52)), COPD (HRAge+Sex 1.56 (95% CI 1.53–1.6), HRPANEL 1.35
(95% CI 1.31–1.39)) or MACE (HRAge+Sex 1.63 (95% CI 1.58–1.69),
HRPANEL 1.4 (95% CI 1.33–1.46)), showed less pronounced, yet
clear, separation of risk trajectories. In regard to T2D, the HRs of
the metabolomic states were externally validated with adjustment
for Age+Sex for all-cause dementia, heart failure, atrial fibrillation,
CHD and COPD (Extended Data Fig. 3b). In contrast, the metabo-
lomic state only marginally modified the risk trajectories for asthma
(HRAge+Sex 1.37 (95% CI 1.3–1.44), HRPANEL 1.09 (95% CI 1.03–1.16))
and cataract (HRAge+Sex 1.22 (95% CI 1.18–1.25), HRPANEL 1.08 (95%
CI 1.05–1.11)).
Discriminative performance translates to clinical utility. While
discrimination is critical, the clinical utility of any risk model
depends on calibration and the choice of adequate thresholds
for interventions. We found all models well calibrated in the UK
Biobank Cohort (see Fig. 4a–c and Supplementary Fig. 1 for details
on all endpoints). UK Biobank37, as one of the largest and most com-
prehensive population cohorts in the world, therefore, allowed us to
estimate clinical utility with high precision over a wide range of clin-
ically reasonable intervention thresholds. However, adequate clini-
cal decision thresholds directly depend on the benefits and harms of
interventions and disease prevalence. We therefore calculated deci-
sion curves38 to estimate the benefit of adding metabolomic infor-
mation to a prediction model (see Fig. 4d–i and Supplementary
Fig. 1 for details on all endpoints). Further, we calculated clini-
cally relevant metrics such as sensitivity, positive predictive value
and positive likelihood ratio over multiple false-positive rates
(Supplementary Table 11)39.
Specifically, we investigated the application of the metabolomic
state in two scenarios. First, as a potentially economical and practi-
cal option, we assessed the combination of the metabolomic state
with Age+Sex and with the less resource-intensive, non-laboratory
predictors of the PANEL set. Second, we combined the metabolomic
state with the entire PANEL set (including all laboratory predictors)
to assess whether there is a net benefit even beyond comprehensive
predictors.
Generally we found that discriminative gains (Fig. 3) translated
to utility gains (see Fig. 4d–i and Supplementary Fig. 1 for details on
all endpoints). The metabolomic state substantially added to age and
sex for most endpoints, and additional non-laboratory predictors
either closed (12 of the 24 endpoints, including T2D, stroke, heart
failure and lung cancer) or narrowed the gap (an additional four
of the 24 endpoints, including dementia, atrial fibrillation and renal
disease) with the comprehensive set of PANEL predictors. The addi-
tion of the metabolomic state to the comprehensive PANEL predic-
tors led to further improvements in the utility for reasonable ranges
of decision thresholds for 11 of the 24 endpoints (most notably T2D,
heart failure and, to a lesser extent, dementia; see Supplementary
Fig. 1 for details on all endpoints and Extended Data Fig. 7 for
NATURE MEDICINE | VOL 28 | NOVEMBER 2022 | 2309–2320 | www.nature.com/naturemedicine 2315
Content courtesy of Springer Nature, terms of use apply. Rights reserved
Articles Nature MediciNe
additional analyses investigating apolipoprotein 4 (APOE4) carrier
status for dementia). Conversely, where there were no improve-
ments in the discriminatory value, no relevant improvements
in clinical utility could be found. These observations were further
reflected in the positive predictive values and positive likelihood
ratios (Supplementary Table 9).
d
75
100
Standardized net benefit (%)Standardized net benefit (%)
Standardized net benefit (%)
Standardized net benefit (%)
T2D Heart failure
0
25
50
0 10 20 30 40 50
Threshold probability (%) Threshold probability (%)
0 5 10 15 20
Observed event rate (%)
T2D
a
10
15
20
0
5
0 5 10 15 20
Predicted risk (%)
0
25
50
75
100
0 10 20 30 40 50
Threshold probability (%) Threshold probability (%) Threshold probability (%)
T2D
0
25
50
75
100
0 2.5 5.0 7.5
Dementia
0
25
50
75
100
0 5 10 15 20
Heart failure
h ig
Age+Sex ASCVD PANEL Age+Sex + MET PANELnoLaboratory + MET Age+Sex versus Age+Sex + MET Age+Sex + MET
versus PANELnoLaboratory
+ MET
Age+Sex ASCVD PANEL Age+Sex + MET PANELnoLaboratory + MET PANEL + MET Calibration line
PANEL versus PANEL + METAg+Sex ASCVD PANEL PANEL + MET
e f
Standardized net benefit (%)
Standardized net benefit (%)
75
100
75
Dementia
100
Threshold probability (%)
0
25
50
0 2.5 5.0 7.5
0
25
50
Observed event rate (%)
Observed event rate (%)
2
3
4
Dementia
5.0
7.5
10.0
Heart failure
b c
Predicted risk (%) Predicted risk (%)
0
1
012345
0
2.5
0 2.5 5.0 7.5 10.0
Fig. 4 | Model calibration and additive predictive value of the metabolomic state translate to potential clinical utility. a–c, Calibration curves for CPH
models, including baseline parameter sets Age+Sex, ASCVD and PANEL, as well as their combinations with the metabolomic state (Age+Sex + MET)
for the endpoints T2D (a), dementia (b) and heart failure (c). d–f, Endpoint-specific net benefit curves standardized by endpoint prevalence, where
horizontal solid gray lines indicate ‘treat none’ and vertical solid gray lines indicate ‘treat all’; T2D (d), dementia (e) and heart failure (f). The
standardized net benefits of sets Age+Sex, ASCVD and PANEL are compared with Age+Sex + MET and additional non-laboratory predictors of PANEL
(PANELnoLaboratory). Green and blue color-filled areas indicate the added benefit of the combination of the metabolomic state and Age+Sex and
PANELnoLaboratory, respectively. g–i, Standardized net benefit curves comparing the performance of PANEL + MET against baselines Age+Sex, ASCVD
and PANEL; T2D (g), dementia (h) and heart failure (i). Decision curves were derived from n = 111,745 (T2D), n = 117,245 (dementia) and n = 113,636
(heart failure) individuals.
NATURE MEDICINE | VOL 28 | NOVEMBER 2022 | 2309–2320 | www.nature.com/naturemedicine
2316
Content courtesy of Springer Nature, terms of use apply. Rights reserved
Articles
Nature MediciNe
Amino acids
BCAAs
AAAs
Fluid balance
Inflammation
FAs
Glyco lysis-
related
metabolites
Ketone bodies
Total lipids
Cholesterol
Free
cholesterol
Cholesterol
esters
Phospholipids
Triglycerides
Other lipids
Lipoprotein
parti cle sizes
Lipoprotein
parti cle
concentrations
Chylomicrons
and extremely
large VLDL
Very larg e VLDL
Large VLDL
Medium VLDL
Small VLDL
Very small VLDL
IDL
Large LDL
Medium LDL
Small LDL
Very larg e HDL
Large HDL
Medium HDL
Small HDL
Apolipoproteins
alA
Gln
Gly
His
Ile
Leu
Total BCAA
Val
Phe
Tyr
Albumin
Creatinine
GlycA
DHA
LA
MUFA
Omega−3
Omega−6
PUFA
SFA
Total−FA
Unsaturation
Citrate
Glucose
Lactate
Pyruvate
Acetate
Acetoacetate
Acetone
bOHbutyrate
HDL−L
LDL−L
Total−L
VLDL−L
Clinical LDL−C
HDL−C
LDL−C
non−HDL−C
Remnant−C
Total−C
VLDL−C
HDL−FC
LDL−FC
Total−FC
VLDL−FC
HDL−CE
LDL−CE
Total−CE
VLDL−CE
HDL−PL
LDL−PL
Total−PL
VLDL−PL
HDL−TG
LDL−TG
Total−TG
VLDL−TG
Cholines
Phosphatidylc
Phosphoglyc
Sphingomyelins
HDL size
LDL size
VLDL size
HDL−P
LDL−P
Total−P
VLDL−P
XXL−VLDL−C
XXL−VLDL−CE
XXL−VLDL−FC
XXL−VLDL−L
XXL−VLDL−P
XXL−VLDL−PL
XXL−VLDL−TG
XL−VLDL−C
XL−VLDL−CE
XL−VLDL−FC
XL−VLDL−L
XL−VLDL−P
LP−LDLV−LX
XL−VLDL−TG
L−VLDL−C
L−VLDL−CE
L−VLDL−FC
L−VLDL−L
L−VLDL−P
L−VLDL−PL
L−VLDL−TG
M−VLDL−C
M−VLDL−CE
M−VLDL−FC
M−VLDL−L
M−VLDL−P
M−VLDL−PL
M−VLDL−TG
S−VLDL−C
S−VLDL−CE
S−VLDL−FC
S−VLDL−L
S−VLDL−P
S−VLDL−PL
S−VLDL−TG
XS−VLDL−C
XS−VLDL−CE
XS−VLDL−FC
XS−VLDL−L
XS−VLDL−P
XS−VLDL−PL
XS−VLDL−TG
IDL−C
IDL−CE
IDL−FC
IDL−L
IDL−P
IDL−PL
IDL−TG
L−LDL−C
L−LDL−CE
L−LDL−FC
L−LDL−L
L−LDL−P
L−LDL−PL
L−LDL−TG
M−LDL−C
M−LDL−CE
M−LDL−FC
M−LDL−L
M−LDL−P
M−LDL−PL
M−LDL−TG
S−LDL−C
S−LDL−CE
S−LDL−FC
S−LDL−L
S−LDL−P
S−LDL−PL
S−LDL−TG
XL−HDL−C
XL−HDL−CE
XL−HDL−FC
XL−HDL−L
XL−HDL−P
XL−HDL−PL
XL−HDL−TG
L−HDL−C
L−HDL−CE
L−HDL−FC
L−HDL−L
L−HDL−P
L−HDL−PL
L−HDL−TG
M−HDL−C
M−HDL−CE
M−HDL−FC
M−HDL−L
M−HDL−P
M−HDL−PL
M−HDL−TG
S−HDL−C
S−HDL−CE
S−HDL−FC
S−HDL−L
S−HDL−P
S−HDL−PL
S−HDL−TG
ApoA1
BopA
T2D
Amino acids
BCAAs
AAA
Fluid balance
Inflammation
FAs
Glyco lysis-
related
metabolites
Ketone bodies
Total lipids
Cholesterol
Free
cholesterol
Cholesterol
esters
Phospholipids
Triglycerides
Other lipids
Lipoprotein
parti cle sizes
Lipoprotein
parti cle
concentrations
Chylomicrons
and extremely
large VLDL
Very larg e VLDL
Large VLDL
Medium VLDL
Small VLDL
Very small VLDL
IDL
Large LDL
Medium LDL
Small LDL
Very larg e HDL
Large HDL
Medium HDL
Small HDL
Apolipoproteins
alA
Gln
Gly
His
Ile
Leu
Total BCAA
Val
Phe
Tyr
Albumin
Creatinine
GlycA
DHA
LA
MUFA
Omega−3
Omega−6
PUFA
SFA
Total−FA
Unsaturation
Citrate
Glucose
Lactate
Pyruvate
Acetate
Acetoacetate
Acetone
bOHbutyrate
HDL−L
LDL−L
Total−L
VLDL−L
Clinical LDL−C
HDL−C
LDL−C
non−HDL−C
Remnant−C
Total−C
VLDL−C
HDL−FC
LDL−FC
Total−FC
VLDL−FC
HDL−CE
LDL−CE
Total−CE
VLDL−CE
HDL−PL
LDL−PL
Total−PL
VLDL−PL
HDL−TG
LDL−TG
Total−TG
VLDL−TG
Cholines
Phosphatidylc
Phosphoglyc
Sphingomyelins
HDL size
LDL size
VLDL size
HDL−P
LDL−P
Total−P
VLDL−P
XXL−VLDL−C
XXL−VLDL−CE
XXL−VLDL−FC
XXL−VLDL−L
XXL−VLDL−P
XXL−VLDL−PL
XXL−VLDL−TG
XL−VLDL−C
XL−VLDL−CE
XL−VLDL−FC
XL−VLDL−L
XL−VLDL−P
LP−LDLV−LX
XL−VLDL−TG
L−VLDL−C
L−VLDL−CE
L−VLDL−FC
L−VLDL−L
L−VLDL−P
L−VLDL−PL
L−VLDL−TG
M−VLDL−C
M−VLDL−CE
M−VLDL−FC
M−VLDL−L
M−VLDL−P
M−VLDL−PL
M−VLDL−TG
S−VLDL−C
S−VLDL−CE
S−VLDL−FC
S−VLDL−L
S−VLDL−P
S−VLDL−PL
S−VLDL−TG
XS−VLDL−C
XS−VLDL−CE
XS−VLDL−FC
XS−VLDL−L
XS−VLDL−P
XS−VLDL−PL
XS−VLDL−TG
IDL−C
IDL−CE
IDL−FC
IDL−L
IDL−P
IDL−PL
IDL−TG
L−LDL−C
L−LDL−CE
L−LDL−FC
L−LDL−L
L−LDL−P
L−LDL−PL
L−LDL−TG
M−LDL−C
M−LDL−CE
M−LDL−FC
M−LDL−L
M−LDL−P
M−LDL−PL
M−LDL−TG
S−LDL−C
S−LDL−CE
S−LDL−FC
S−LDL−L
S−LDL−P
S−LDL−PL
S−LDL−TG
XL−HDL−C
XL−HDL−CE
XL−HDL−FC
XL−HDL−L
XL−HDL−P
XL−HDL−PL
XL−HDL−TG
L−HDL−C
L−HDL−CE
L−HDL−FC
L−HDL−L
L−HDL−P
L−HDL−PL
L−HDL−TG
M−HDL−C
M−HDL−CE
M−HDL−FC
M−HDL−L
M−HDL−P
M−HDL−PL
M−HDL−TG
S−HDL−C
S−HDL−CE
S−HDL−FC
S−HDL−L
S−HDL−P
S−HDL−PL
S−HDL−TG
ApoA1
BopA
Dementia
+ Positive contribution
– Negative contribution
b
c
a
Scaled global absolute SHAP value
Amino acids
BCAAs
AAAs
Fluid balance
Inflammation
FAs
Glycolysis-related
metabolites
Ketone bodies
Cholesterol
Free cholesterol
Cholesteryl esters
Triglycerides
Other lipids
Lipoprotein
particle sizes
Chylomicrons and
extremely large
VLDL
Very large VLDL
Large VLDL
Medium VLDL
Small VLDL
Very small VLDL
Large LDL
Medium LDL
Small LDL
IDL
Very large HDL
Large HDL
Medium HDL
Small HDL
Apolipoproteins
T2D
AAA
Lung cancer
Heart failure
Cerebral stroke
Atrial fibrillation
MACE
CHD
PAD
Liver disease
Renal disease
Cataract
Dementia
COPD
Parkinson’s
Ven. thrombosis
Rectal cancer
Colon cancer
Asthma
Fracture
Skin cancer
Glaucoma
Prostate cancer
Breast cancer
His
Gly
Gln
Ala
Val
Total BCAA
Leu
Ile
Tyr
Phe
Creatinine
Albumin
GlycA
Unsaturation
Total–FA
SFA
Omega–3
MUFA
LA
DHA
Pyruvate
Lactate
Glucose
Citrate
bOHbutyrate
Acetone
Acetoacetate
VLDL–C
Total–C
Remnant–C
VLDL–FC
Total–FC
VLDL–CE
LDL–TG
HDL–TG
Sphingomyelins
Cholines
LDL size
XXL–VLDL–CE
XXL–VLDL–C
XL–VLDL–FC
XL–VLDL–CE
L–VLDL–TG
L–VLDL–L
M–VLDL–L
M–VLDL–CE
M–VLDL–C
S–VLDL–TG
S–VLDL–P
XS–VLDL–PL
XS–VLDL–P
XS–VLDL–FC
L–LDL–TG
L–LDL–C
M–LDL–TG
M–LDL–FC
S–LDL–TG
S–LDL–PL
S–LDL–FC
S–LDL–CE
S–LDL–C
IDL–TG
IDL–PL
IDL–P
IDL–FC
XL–HDL–FC
L–HDL–L
L–HDL–CE
L–HDL–C
M–HDL–TG
M–HDL–PL
M–HDL–L
S–HDL–P
S–HDL–FC
ApoB
0 0.5 1.0
–1.0 –0.5 0 0.5 1.0
Normalized
metabolite value
Low High
Fig. 5 | Analysis of the metabolomic state informs on metabolite profiles associated with disease risk. a, Heatmap showing the importance of metabolites
in regard to the estimated metabolomic states, represented by absolute global SHAP value estimates per endpoint for the 75 globally most important
metabolites. Endpoints are sorted by the discriminative performance of the metabolomic state (left to right; Fig. 3a). b, Global metabolite attributions
for T2D; individual attributions are aggregated by percentiles and each dot indicates one percentile. The more distant a dot from the circular baseline,
the stronger the absolute attribution for that percentile. Deviations toward the center and periphery represent negative and positive contributions,
respectively, to the metabolomic state. Colors indicate the metabolite’s mean plasma value. c, Global metabolite attributions for all-cause dementia. IDL,
intermediate-density lipoprotein.
NATURE MEDICINE | VOL 28 | NOVEMBER 2022 | 2309–2320 | www.nature.com/naturemedicine 2317
Content courtesy of Springer Nature, terms of use apply. Rights reserved
Articles Nature MediciNe
Identification of disease-specific metabolite profiles. A require-
ment for the adoption of neural networks in medicine is explain-
ability. While neural networks are not inherently interpretable,
methods have been developed to overcome this challenge40. To
identify which metabolites most affect disease risk, we approxi-
mated Shapely additive explanation (SHAP) values41 for all inves-
tigated diseases. Generally, the larger the absolute SHAP value the
more important a metabolite for an individual prediction. Based on
the direction of the effect of a metabolite’s contribution, increasing
or decreasing the predicted risk, SHAP can take a positive or nega-
tive value.
To understand individual metabolites in the context of the
24 investigated diseases, we investigated global metabolite attribu-
tions, the sum of absolute SHAP values per metabolite and disease
(Fig. 5a and Extended Data Fig. 8). We found that most high-impact
metabolites were linked to multiple diseases: plasma levels of metab-
olites with consistently high contribution included the amino acids
glutamine, glycine and tyrosine, metabolites related to carbohydrate
metabolism, albumin, the kidney function marker creatinine, glyco-
protein acetylation (GlycA) and the ketone bodies acetone and ace-
toacetate. Further implicated were fatty acids (FA) such as linoleic
acid (LA) and multiple lipoprotein components, including free cho-
lesterol in very large high-density lipoprotein (VHDL), triglycerides
in large low-density lipoprotein (LDL), phospholipids in small LDL
and sphingomyelins. In addition to shared metabolite profiles, we
pinpointed marked associations of creatinine with AAA, glucose
with T2D and GlycA with lung cancer and COPD. For diseases
with a high discriminatory value for metabolomic state, predicted
metabolite contributions were considerably higher than for diseases
with little discriminatory metabolomic information (Fig. 5a).
Subsequently we focused on T2D (Fig. 5b) and all-cause dementia
(Fig. 5c), two diseases with strong metabolomic contributions over
the comprehensive clinical predictors and indications of clinical util-
ity (see above). Metabolites related to carbohydrate metabolism, such
as glucose and lactate, dominated the predicted metabolomic state
of T2D in our model (Fig. 5b). In line with earlier findings5,19,42, we
observed contributions of amino acids, ketone bodies, lipids and FAs
as well as creatinine and albumin. We confirmed that higher plasma
levels of FAs, docosahexaenoic acid (DHA) and LA were associated
with lower risk43,44. Further, we observed a distinct contribution
of lipid content across the whole density gradient of lipoproteins,
including a high triglyceride content in LDL and IDL or free cho-
lesterol content in very small very-low-density lipoprotein (VLDL)
and HDL. For all-cause dementia, we identified creatinine, albumin
and the amino acids glutamine, leucine and tyrosine as predominant
contributors to predicted risk (Fig. 5c). In line with earlier findings8,45,
we observed a notable role of FAs such as LA and monounsaturated
and saturated FAs, as well as a protective effect of branched-chain
amino acids (BCAAs). Our results further implicate associations
of glucose, ketone bodies acetate, acetoacetate and acetone, and
beta-hydroxybutyrate. Finally we found several lipoproteins to be
associated, most notably free cholesterol in very large HDL and cho-
lesterylester in extremely large VLDL. Comprehensive data for all
investigated endpoints, including the most important metabolites
and disease-specific attribution profiles, can be found in Extended
Data Fig. 8, Supplementary Table 12 and Supplementary Fig. 2).
Computation of SHAP values also allowed us to derive individ-
ual risk attribution profiles for individual participants and diseases,
informing on the impact of single metabolites on a given prediction.
We visualize the attribution profiles for T2D in two-dimensional
uniform manifold approximation and projection (UMAP)46 space
(Extended Data Fig. 9), which is resolved by the estimated impor-
tance of glucose (that is, SHAP values assigned to glucose regard-
ing the predicted risk for T2D; Extended Data Fig. 9a). While most
high-risk individuals (top 1% metabolomic state) are located at
coordinates with strong glucose attribution, we found high-risk
individuals scattered over the entire attribution space (Extended
Data Fig. 9b). Interestingly, the attribution profiles of high-risk
individuals were not consistently dominated by glucose but rather
by, for instance, low levels of albumin, LA, DHA, histidine and gly-
cine (Extended Data Fig. 9c). This observation is further reflected in
NMR metabolite concentrations, because we found substantial dif-
ferences in the concentrations of glucose, LA, FAs and triglycerides
when comparing the metabolite distributions of individuals in the
area with the strongest glucose attribution with those of individu-
als in two spatially distinct, high-risk UMAP areas (Extended Data
Fig. 10).
Discussion
The assessment of risk is a critical component of disease prevention.
However, comprehensive risk assessment often requires the care-
ful acquisition of predictors, one disease at a time. Thus, for each
disease-specific risk score, the resources (time and cost) required for
the collection can severely limit adoption and utility47. Interestingly,
many common diseases involve metabolic alterations and human
blood metabolomic patterns contain rich systemic information on
the underlying physiology9–11,20,21. While individual metabolites have
long been linked to disease risk, systemic information from blood
metabolomics could inform on multiple diseases simultaneously.
Importantly, in recent years, assays such as 1H-NMR spectroscopy
have matured and allowed the assessment of serum metabolite
information robustly at comparatively low cost13,14. However, the
potential of metabolomic profile as a single-domain, multidisease
assay in primary care has not been investigated thus far.
We have assessed the potential of NMR-derived metabolo-
mic profiles as a tool for individualized prediction of onset across
24 common diseases. With >1.4 million person-years of follow-up,
we leveraged the systemic information in metabolomic profiles to
derive integrative metabolomic states for many diseases simultane-
ously. We found the metabolomic states to be predictive for all but
one of the investigated diseases and externally validated these find-
ings in four independent cohorts for available endpoints. Further,
we investigated the predictive value beyond clinical variables and
identified a subset of endpoints with potential clinical utility.
Finally, we examined metabolite attributions confirming a multi-
tude of disease-associated metabolites and a shared metabolomic
background of common diseases.
Importantly, we found that the predictive information of the
metabolomic state matched established clinical variables for many of
the investigated endpoints. In line with previous reports on NMR–
metabolite associations, we confirm that metabolomic profiles are
highly predictive for, for example, T2D19, dementia8 and cardiovas-
cular diseases6,11,17 such as CHD and heart failure48. Generally, the
additional predictive information decreased over comprehensive
clinical predictors, indicating that substantial parts of the metabo-
lomic state’s discriminatory information are shared with established
clinical predictors. However, for multiple endpoints, including
T2D, all-cause dementia and heart failure, the metabolomic state
contained complementary information that added predictive value
even over comprehensive laboratory measurements. These findings
largely translate into potential clinical utility for NMR-based metab-
olomic profiling, both as a replacement for comprehensive labora-
tory examinations and as an additional source of discriminatory
information to refine comprehensive risk assessments for multiple
diseases simultaneously.
Calculation of attributions for each individual allowed us to
assess how differences in the metabolomic profile affect disease
risk. We confirmed the role of metabolites such as albumin and
creatinine, which have previously been associated with all-cause
and disease-specific mortality11,16 and are already part of routine
care49,50. Further, we confirmed the role of LA, tyrosine, glycine
and cholesterylesters in extremely large VLDL in multiple diseases,
NATURE MEDICINE | VOL 28 | NOVEMBER 2022 | 2309–2320 | www.nature.com/naturemedicine
2318
Content courtesy of Springer Nature, terms of use apply. Rights reserved
Articles
Nature MediciNe
further supporting metabolomic multidisease-spanning informa-
tion21. Dissecting disease-specific attribution profiles, we found
that metabolite attributions reflect metabolite–disease associa-
tions previously reported in the literature. In the case of T2D, we
confirmed the associations between disease risk and metabolites
beyond glucose. Specifically, our model captured the positive asso-
ciation between high levels of glycoprotein acetyls, BCAAs, lactate
and FAs (both monounsaturated and saturated) and the protective
role of metabolites such as LA or glycine5,19. In the attribution pro-
file of dementia we replicated associations with BCAAs, including
leucine and valine, and with FAs, most notably LA8,45. In addition,
the associations of GlycAs with cardiovascular disease, T2D, COPD
and lung cancer51,52 are reflected in the attributions. Consequently,
our metabolomic state model learns systemic information in
NMR-derived metabolomic profiles based on established shared
and highly specific metabolite–disease associations.
In our perspective, 1H-NMR metabolomics profiling is an attrac-
tive candidate for a single-domain, multidisease assay. Because many
countries already recommend regular check-ups entailing blood
tests in the prevention of selected common diseases53, our results
indicate the potential of NMR metabolomic profiling in combina-
tion with simple demographic, but also with comprehensive labora-
tory predictors to estimate disease risk. In addition, metabolomic
risk profiles could be of potential value in the guidance of pharma-
cological and lifestyle interventions. This is especially relevant for
diseases such as T2D, where interventions on modifiable risk factors
have been shown to delay disease onset54 and prevent subsequent
comorbidities55,56. Similarly, the Lancet 2020 commission suggested
that up to 40% of worldwide dementia may be preventable by inter-
ventions on modifiable risk factors57. This is particularly compelling
because today’s pharmacological treatment options for dementia are
scarce. However, the efficacy of various lifestyle interventions58,59 is
disputed, calling for further experimental investigation.
Before application in routine care, substantial challenges
remain. While the 1H-NMR assay is robust and cheaper than
mass-spectrometry-based alternatives, sensitivity is lower. Also, cur-
rent metabolite coverage is relatively narrow and lipid focused13,14,60.
Although a future expansion of metabolite coverage is expected, it
presents a limitation for clinical utility to date. Further, downstream
quantification from raw NMR spectra needs to be harmonized for
the reliable application of multivariable prediction models. While
our study population is more healthy and less deprived than the
general UK population37, the results of external validation in four
independent cohorts indicate general transferability of metabolo-
mic states. However, the scope of validation was limited by the avail-
able endpoint information, constraining the replication to a subset
of seven endpoints. In light of these limitations, we recommend
careful scrutinization before application of the metabolomic state
model beyond the validated conditions or in specific populations
outside the research context. Ultimately, a broad rollout of NMR
metabolomics for clinical care requires multiple logistical questions
to be addressed, including both sample processing and transport.
Taken together, our work demonstrates the potential and limita-
tions of NMR-derived metabolomic profiles as a multidisease assay
to inform on the risk of many common diseases simultaneously.
Online content
Any methods, additional references, Nature Research report-
ing summaries, source data, extended data, supplementary infor-
mation, acknowledgements, peer review information; details of
author contributions and competing interests; and statements of
data and code availability are available at https://doi.org/10.1038/
s41591-022-01980-3.
Received: 2 November 2021; Accepted: 28 July 2022;
Published online: 22 September 2022
References
1. WHO CVD Risk Chart Working Group. World Health Organization
cardiovascular disease risk charts: revised models to estimate risk in 21 global
regions. Lancet Glob. Health 7, e1332–e1345 (2019).
2. A and B recommendations. U.S. Preventive Services Task Force https://www.
uspreventiveservicestaskforce.org/uspstf/recommendation-topics/
uspstf-and-b-recommendations (2022).
3. Go David, C. et al. 2013 ACC/AHA Guideline on the Assessment of
Cardiovascular Risk. Circulation 129, S49–S73 (2014).
4. Würtz, P. et al. Circulating metabolite predictors of glycemia in middle-aged
men and women. Diabetes Care 35, 1749–1756 (2012).
5. Mahendran, Y. et al. Association of ketone body levels with hyperglycemia
and type 2 diabetes in 9,398 Finnish men. Diabetes 62, 3618–3626 (2013).
6. Holmes, M. V. et al. Lipids, lipoproteins, and metabolites and risk of
myocardial infarction and stroke. J. Am. Coll. Cardiol. 71, 620–632 (2018).
7. Lécuyer, L. et al. NMR metabolomic signatures reveal predictive plasma
metabolites associated with long-term risk of developing breast cancer. Int. J.
Epidemiol. 47, 484–494 (2018).
8. Tynkkynen, J. et al. Association of branched-chain amino acids and other
circulating metabolites with risk of incident dementia and Alzheimer’s
disease: a prospective study in eight cohorts. Alzheimers Dement. 14, 723–733
(2018).
9. Ahadi, S. et al. Personal aging markers and ageotypes revealed by deep
longitudinal proling. Nat. Med. 26, 83–90 (2020).
10. Schüssler-Fiorenza Rose, S. M. et al. A longitudinal big data approach for
precision health. Nat. Med. 25, 792–804 (2019).
11. Deelen, J. et al. A metabolic prole of all-cause mortality risk identied in an
observational study of 44,168 individuals. Nat. Commun. 10, 3346 (2019).
12. Markley, J. L. et al. e future of NMR-based metabolomics. Curr. Opin.
Biotechnol. 43, 34–40 (2017).
13. Soininen, P., Kangas, A. J., Würtz, P., Suna, T. & Ala-Korpela, M. Quantitative
serum nuclear magnetic resonance metabolomics in cardiovascular
epidemiology and genetics. Circ. Cardiovasc. Genet. 8, 192–206 (2015).
14. Würtz, P. et al. Quantitative serum nuclear magnetic resonance metabolomics
in large-scale epidemiology: a primer on -omic technologies. Am. J.
Epidemiol. 186, 1084–1096 (2017).
15. Ala-Korpela, M., Zhao, S., Järvelin, M.-R., Mäkinen, V.-P. & Ohukainen, P.
Apt interpretation of comprehensive lipoprotein data in large-scale
epidemiology: disclosure of fundamental structural and metabolic
relationships. Int. J. Epidemiol. 51, 996–1011 (2021)
16. Fischer, K. et al. Biomarker proling by nuclear magnetic resonance
spectroscopy for the prediction of all-cause mortality: an observational study
of 17,345 persons. PLoS Med. 11, e1001606 (2014).
17. Würtz, P. et al. Metabolite proling and cardiovascular event risk: a
prospective study of 3 population-based cohorts. Circulation 131, 774–785
(2015).
18. Fizelova, M. et al. Associations of multiple lipoprotein and apolipoprotein
measures with worsening of glycemia and incident type 2 diabetes in 6607
non-diabetic Finnish men. Atherosclerosis 240, 272–277 (2015).
19. Ahola-Olli, A. V. et al. Circulating metabolites and the risk of type 2 diabetes:
a prospective study of 11,896 young adults from four Finnish cohorts.
Diabetologia 62, 2298–2309 (2019).
20. Julkunen, H., Cichońska, A., Slagboom, P. E. & Würtz, P., Nightingale Health
UK Biobank Initiative. Metabolic biomarker proling for identication of
susceptibility to severe pneumonia and COVID-19 in the general population.
eLife 10, e63033 (2021).
21. Pietzner, M. et al. Plasma metabolites to prole pathways in
noncommunicable disease multimorbidity. Nat. Med. 27, 471–479 (2021).
22. Bycro, C. et al. e UK Biobank resource with deep phenotyping and
genomic data. Nature 562, 203–209 (2018).
23. Cox, D. R. Regression models and life-tables. J. R. Stat. Soc. Ser. B Stat.
Methodol. 34, 187–202 (1972).
24. Marmot, M. & Brunner, E. Cohort prole: the Whitehall II study. Int. J.
Epidemiol. 34, 251–256 (2005).
25. Ikram, M. A. et al. e Rotterdam Study: 2018 update on objectives, design
and main results. Eur. J. Epidemiol. 32, 807–850 (2017).
26. Schoenmaker, M. et al. Evidence of genetic enrichment for exceptional
survival using a family approach: the Leiden Longevity Study. Eur. J. Hum.
Genet. 14, 79–84 (2006).
27. Shepherd, J. et al. Pravastatin in elderly individuals at risk of vascular disease
(PROSPER): a randomised controlled trial. Lancet 360, 1623–1630 (2002).
28. Sudlow, C. et al. UK Biobank: an open access resource for identifying the
causes of a wide range of complex diseases of middle and old age. PLoS Med.
12, e1001779 (2015).
29. Qiu, C. et al. Association of blood pressure and hypertension with the risk
of Parkinson disease: the National FINRISK Study. Hypertension 57,
1094–1100 (2011).
30. de Bruijn, R. F. A. G. & Ikram, M. A. Cardiovascular risk factors and future
risk of Alzheimer’s disease. BMC Med. 12, 130 (2014).
NATURE MEDICINE | VOL 28 | NOVEMBER 2022 | 2309–2320 | www.nature.com/naturemedicine 2319
Content courtesy of Springer Nature, terms of use apply. Rights reserved
Articles Nature MediciNe
31. Johnson, C. B., Davis, M. K., Law, A. & Sulpher, J. Shared risk factors for
cardiovascular disease and cancer: implications for preventive health and
clinical care in oncology patients. Can. J. Cardiol. 32, 900–907 (2016).
32. Lindström, J. & Tuomilehto, J. e diabetes risk score: a practical tool to
predict type 2 diabetes risk. Diabetes Care 26, 725–731 (2003).
33. Sindi, S. et al. e CAIDE Dementia Risk Score App: the development of an
evidence-based mobile application to predict the risk of dementia. Alzheimers
Dement. 1, 328–333 (2015).
34. van der Velde, M. et al. Screening for albuminuria identies individuals at
increased renal risk. J. Am. Soc. Nephrol. 20, 852–862 (2009).
35. Mars, N. et al. Polygenic and clinical risk scores and their impact on age at
onset and prediction of cardiometabolic diseases and common cancers. Nat.
Med. 26, 549–557 (2020).
36. Huang, C. et al. Performance metrics for the comparative analysis of clinical
risk prediction models employing machine learning. Circ. Cardiovasc. Qual.
Outcomes 14, e007526 (2021).
37. Fry, A. et al. Comparison of sociodemographic and health-related
characteristics of UK Biobank participants with those of the general
population. Am. J. Epidemiol. 186, 1026–1034 (2017).
38. Vickers, A. J. & Elkin, E. B. Decision curve analysis: a novel method for
evaluating prediction models. Med. Decis. Making 26, 565–574 (2006).
39. Hingorani, A. D. et al. Polygenic scores in disease prediction: evaluation
using the relevant performance metrics. Preprint at medRxiv https://doi.
org/10.1101/2022.02.18.22271049 (2022).
40. Gilpin, L. H. et al. Explaining explanations: an overview of interpretability of
machine learning. Preprint at https://arxiv.org/abs/1806.00069v3 (2018).
41. Lundberg, S. & Lee, S.-I. A unied approach to interpreting model
predictions. Preprint at https://arxiv.org/abs/1705.07874v2 (2017).
42. Mahendran, Y. et al. Glycerol and fatty acids in serum predict the
development of hyperglycemia and type 2 diabetes in Finnish men. Diabetes
Care 36, 3732–3738 (2013).
43. Wu, J. H. Y. et al. Omega-6 fatty acid biomarkers and incident type 2
diabetes: pooled analysis of individual-level data for 39 740 adults from 20
prospective cohort studies. Lancet Diabetes Endocrinol. 5, 965–974 (2017).
44. Virtanen, J. K., Mursu, J., Voutilainen, S., Uusitupa, M. & Tuomainen, T.-P.
Serum omega-3 polyunsaturated fatty acids and risk of incident type 2
diabetes in men: the Kuopio Ischemic Heart Disease Risk Factor study.
Diabetes Care 37, 189–196 (2014).
45. van der Lee, S. J. et al. Circulating metabolites and general cognitive ability
and dementia: evidence from 11 cohort studies. Alzheimers Dement. 14,
707–722 (2018).
46. McInnes, L., Healy, J., Saul, N. & Großberger, L. UMAP: Uniform Manifold
Approximation and Projection. J. Open Source Sow. 3, 861 (2018).
47. Steyerberg, E. W. et al. Prognosis Research Strategy (PROGRESS) 3:
prognostic model research. PLoS Med. 10, e1001381 (2013).
48. Delles, C. et al. Nuclear magnetic resonance-based metabolomics identies
phenylalanine as a novel predictor of incident heart failure hospitalisation:
results from PROSPER and FINRISK 1997. Eur. J. Heart Fail. 20, 663–673
(2018).
49. Wannamethee, S. G., Shaper, A. G. & Perry, I. J. Serum creatinine
concentration and risk of cardiovascular disease: a possible marker for
increased risk of stroke. Stroke 28, 557–563 (1997).
50. Ronit, A. et al. Plasma albumin and incident cardiovascular disease: results
From the CGPS and an updated meta-analysis. Arterioscler. romb. Vasc.
Biol. 40, 473–482 (2020).
51. Kettunen, J. et al. Biomarker glycoprotein acetyls is associated with
the risk of a wide spectrum of incident diseases and straties mortality
risk in angiography patients. Circ. Genom. Precis. Med. 11, e002234
(2018).
52. Komaromy, A., Reider, B., Jarvas, G. & Guttman, A. Glycoprotein biomarkers
and analysis in chronic obstructive pulmonary disease and lung cancer with
special focus on serum immunoglobulin G. Clin. Chim. Acta 506, 204–213
(2020).
53. NHS Health Check. NHS https://www.nhs.uk/conditions/nhs-health-check/
(accessed 23 August 2022).
54. Balk, E. M. et al. Combined diet and physical activity promotion programs to
prevent type 2 diabetes among persons at increased risk: a systematic review
for the Community Preventive Services Task Force. Ann. Intern. Med. 163,
437–451 (2015).
55. Yusuf, S. et al. Eect of potentially modiable risk factors associated with
myocardial infarction in 52 countries (the INTERHEART study): case-control
study. Lancet 364, 937–952 (2004).
56. Patel, S. A., Winkel, M., Ali, M. K., Narayan, K. M. V. & Mehta, N. K.
Cardiovascular mortality associated with 5 leading risk factors: national and
state preventable fractions estimated from survey data. Ann. Intern. Med. 163,
245–253 (2015).
57. Livingston, G. et al. Dementia prevention, intervention, and care: 2020 report
of the Lancet Commission. Lancet 396, 413–446 (2020).
58. Silarova, B. et al. Eect of communicating phenotypic and genetic risk of
coronary heart disease alongside web-based lifestyle advice: the INFORM
randomised controlled trial. Heart 105, 982–989 (2019).
59. Ngandu, T. et al. e eect of adherence on cognition in a multidomain
lifestyle intervention (FINGER). Alzheimers Dement. https://doi.org/10.1002/
alz.12492 (2021).
60. Emwas, A.-H. M. e strengths and weaknesses of NMR spectroscopy and
mass spectrometry with particular focus on metabolomics research. Methods
Mol. Biol. 1277, 161–193 (2015).
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in
published maps and institutional affiliations.
Open Access This article is licensed under a Creative Commons
Attribution 4.0 International License, which permits use, sharing, adap-
tation, distribution and reproduction in any medium or format, as long
as you give appropriate credit to the original author(s) and the source, provide a link to
the Creative Commons license, and indicate if changes were made. The images or other
third party material in this article are included in the article’s Creative Commons license,
unless indicated otherwise in a credit line to the material. If material is not included in
the article’s Creative Commons license and your intended use is not permitted by statu-
tory regulation or exceeds the permitted use, you will need to obtain permission directly
from the copyright holder. To view a copy of this license, visit http://creativecommons.
org/licenses/by/4.0/.
© The Author(s) 2022
NATURE MEDICINE | VOL 28 | NOVEMBER 2022 | 2309–2320 | www.nature.com/naturemedicine
2320
Content courtesy of Springer Nature, terms of use apply. Rights reserved
Articles
Nature MediciNe
Methods
Data source and endpoint denition. We use data from the UK Biobank cohort,
a sample of the UK’s general population. Participants were enrolled from 2006
to 2010 in 22 recruitment centers across the United Kingdom; the follow-up is
ongoing. e UK Biobank provides NMR metabolomics measured at recruitment
for a subset of individuals: 63,903 women and 54,078 men aged 37–73 years at the
time of baseline assessment.
Details on the characteristics of the external validation cohorts are provided in
Whitehall II Cohort, Rotterdam Study, Leiden Longevity Study and PROspective
Study of Pravastatin in the Elderly at Risk. We investigated a set of 24 endpoints,
each defined by the earliest occurrence in primary care, hospital episode statistics
or death records. Endpoints were adapted from an earlier study21 and defined
by ICD10 codes (Supplementary Table 1), and patients with previous disease
were excluded for each endpoint. In the case of cardiovascular endpoints we also
excluded patients with lipid-lowering therapy records. In addition, we analyzed
only men or only women for predominantly sex-specific diseases such as prostate
and breast cancer.
The study adhered to the transparent reporting of a multivariable prediction
model for individual prognosis or diagnosis (TRIPOD) statement for reporting61.
The complete checklist can be found in Supplementary Note 1.
Predictor selection and extraction. We investigated three sets of clinical predictor
sets—Age+Sex, ASCVD and PANEL. An overview of the predictors and their
use in the respective covariate sets is presented in Fig. 1d. The NMR assay covers
168 metabolites, from multiple amino acids to lipids, lipoproteins, cholesterol
subtypes and inflammation markers. While the NMR assay further includes
81 percentage ratios derived from combinations of the 168 original measures,
these were not included in the analysis. Basic demographic information was
extracted from primary care records and matched with data collected at the study’s
recruitment interview. Lifestyle information was extracted from the questionnaire
completed at recruitment. Physical measurements and laboratory measures
were taken at recruitment. Pre-existing medical conditions were extracted from
the questionnaire, interview at recruitment, primary care records and hospital
episode statistics. Medications were extracted from the recruitment interview.
Cardiovascular predictors were selected based on ESC- and AHA-recommended
cardiovascular risk scores for primary prevention, the AHA–ASCVD score3 and
the ESC–SCORE2 (ref. 62). For the PANEL predictor set we included additional
predictors from CAIDE33 and FINDRISC32 scores and comprehensive information
on lifestyle, demographics, physical measurements and laboratory values available
in the primary care setting. Because genotyping is currently not commonly
available in primary care, we decided to omit the APOE4 status in the primary
analysis. A dedicated analysis, including APOE4 carrier status for all-cause
dementia, can be found in Extended Data Fig. 6. A list of all clinical predictors
applied in this study is presented in Supplementary Table 2 and a list of all
metabolomic predictors in Supplementary Table 3.
Dataset partitions and imputation. For model development and testing, we
split the dataset into 22 spatially separated partitions based on the location of
the assessment center at recruitment as previously established63. We analyzed
the data in 22-fold nested cross-validation, setting aside one of the spatially
separated partitions as a test set, aggregating the remaining partitions and
randomly selecting 10% of the aggregated data for the validation set. Within
each of the 22 cross-validation loops, the individual test set (that is, the spatially
separated partition) remained untouched throughout model development and the
validation set was used to validate the fitting progress and checkpoint selection. All
22 obtained models were then evaluated on their respective test sets. We assumed
that missing data occurred at random and performed multiple imputations using
chained equations with random forests64. Continuous variables were standardized;
Categorical variables were one-hot encoded. Imputation models were fitted on the
training sets and applied to the respective validation and test sets.
Metabolomic state model. The metabolomic state model is a residual
neural network simultaneously predicting the metabolomic state for each
of the 24 endpoints. The model consists of a shared network and smaller
endpoint-specific head networks. The shared neural network comprises three
fully connected linear layers, each with batch normalization, dropout65 of 0.3 and
sigmoid-weighted linear units (SiLU)66 activations with 256, 256 and 512 nodes. It
outputs a representation of size 512, which is passed on to the endpoint-specific
residual head networks. Thereby, each of the 24 residual head networks takes
two inputs: the shared representation learned by the shared network and the
original 168 metabolomic markers. Each residual head network consists of a small
256-, 128- or 32-node multilayer perceptron (MLP) with a dropout of 0.6, batch
normalization and SiLU activations that transform the shared representation,
and a skip-connection67 network of 128, 128 and 32 nodes transforming the
168 metabolomic markers. The outputs of both networks are subsequently
added in a skip-connection and fed through another two-layer, fully connected
network of 128 and 128 nodes with a dropout of 0.6, batch normalization and
SiLU activations before the scalar metabolomic state is computed through a final
single-output linear layer with identity activation. For each endpoint, and thus for
each metabolomic state, we individually calculate an adapted proportional hazards
loss68, excluding prevalent events endpoint specifically. The individual losses are
averaged and then summed to derive the final loss of the metabolomic state model.
After architecture development, a hyperparameter search is run on training and
validation splits of partition zero as random search over a constrained parameter
space tuning batch size, initial learning rate, number of nodes in the layers of the
endpoint heads and size of the output vector of the shared network. The final
models are trained with batch size 1,024 for a maximum of 100 epochs using the
Adam optimizer69 with default parameters, stochastic weight averaging, a learning
rate of 0.001 and early stopping tracking of the performance on each partition’s
validation set. We further apply a multistep learning rate schedule with gamma
0.1 and steps at 20, 30 and 40 epochs. We implement the metabolomic state model
model in Python v.3.7 using PyTorch v.1.7 (ref. 70) and PyTorch-lightning v.1.4.
Survival analysis and metabolomic state integration. We fitted CPH models23
to derive risk predictions for the individual endpoints. Specifically, for each
endpoint we developed models on seven distinct covariate sets: first, only the
learned metabolomic state; second, the three clinical predictor sets age and sex,
cardiovascular predictors and the comprehensive PANEL (Table 1, Fig. 1d and
Predictor selection and extraction); and third, clinical predictors with the added
metabolomic states for the respective endpoint. Model development was repeated
independently for each assessment center and thus, for each cross-validation
split, models were trained on the respective training set and checkpoints for the
metabolomic state model were selected on the respective validation set. For the final
evaluation, predictions made on the respective test sets were aggregated. Harrell’s
C-index was calculated with the Python package lifelines71 by bootstrapping both
the aggregated test set and individual assessment centers. Statistical inferences about
model differences were based on the distribution of bootstrapped differences in
the C-index; performances were considered significantly different when the 95%
CIs of the performance deltas did not overlap with 0. CPH models were fitted with
CoxPHFitter from the Python package lifelines71, with default parameters and
step size of 0.5 and 0.1 to facilitate model convergence. To estimate risk trajectory
based on the metabolomic state, partial metabolomics effects were calculated using
a custom adaptation of lifelines CoxPHFitter’s plot_partial_effects_on_outcome
method, fixing all other predictors to their central values. CIs for all statistical
analyses were calculated with >1,000 bootstrapping iterations. All statistical analyses
were performed in R v.4.0.2 (ref. 72).
Feature attribution estimates. SHAP values41,73 were calculated to estimate
feature attribution for each endpoint and model individually. SHAP values are a
combination of game-theoretically optimal Shapley values, which determine the
estimated average marginal contribution of each feature for a prediction with local
additivity41,73. Because computation time of exact SHAP values grows exponentially
with an increasing number of features, we resort to an approximation of SHAP
values: DeepSHAP, an adaptation of the DeepLIFT74 method. Importantly, the sum
of the approximated SHAP values amounts to the difference between the expected
model prediction on a given set of background samples and the prediction for an
observed sample. Calculations were performed using the DeepExplainer method
implemented in v.0.39 of the SHAP package75. After calculation of per-sample
attributions for each metabolite and endpoint, attributions were aggregated per
endpoint to derive a global metabolite-specific set of attributions. We identified
important attributes based on the top and bottom 1% percentile borders of the
SHAP value distribution over all attributions.
Individual metabolite attribution profiles. Computation of SHAP values
(Feature attribution estimates) enabled the derivation of attribution profiles for
each individual and disease, informing on the specific contribution of metabolites
to individual risk. Individual high-impact metabolites were defined by the
top and bottom 1% percentiles of the metabolite SHAP distribution (that is,
SHAP ∉ (−0.2,0.2)). To assess the space of individual attribution profiles, UMAP46
for dimension reduction was fitted on the entire set of SHAP values for each
endpoint individually. The UMAP projection allows assessment of the complex,
high-dimensional manifold of attribution values in two-dimensional space.
UMAPs were fitted using the UMAP Python package76 and default parameters. For
visualization of UMAP space, 41 unconnected outliers of 117,981 total observations
were excluded.
Replication in independent cohorts. The models fitted in the UK Biobank were
exported via ONNX77, and calculation of metabolomic states was replicated in
the Whitehall II Cohort24, the Rotterdam Study25, the Leiden Longevity Study26
and the PROspective Study of Pravastatin in the Elderly at Risk27,78 (Fig. 1c and
Supplementary Table 4). In consideration of available predictors and endpoints,
CPH models were fitted and evaluated as described in Survival analysis and
metabolomic state integration. The ONNX weights of the model, as well as the
normalization pipeline for the NMR data as fitted on the UK Biobank, are available
through our GitHub repository (Code availability).
Whitehall II Cohort. The Whitehall II Cohort (WHII) is an ongoing prospective
cohort study of adults, consisting of 10,308 individuals (3,413 women and
NATURE MEDICINE | www.nature.com/naturemedicine
Content courtesy of Springer Nature, terms of use apply. Rights reserved
Articles Nature MediciNe
6,895 men) recruited at age 35–55 years24. At the time of recruitment (1985–1988),
all study participants were working in the London offices of 20 Whitehall
departments. Participants have been followed up regularly over the years, with
questionnaires and self-examination conducted every 5 years. NMR profiling was
performed from serum samples between 1997 and 1999.
Rotterdam Study. The Rotterdam Study (RS) is a prospective, population-based
cohort study25 with the aim of determining the occurrence of common diseases
in elderly people. Baseline examination took place in 1990, with approximately
7,983 persons aged 55 years and older undergoing a home interview and extensive
physical examination. Follow-up visits took place every 3–4 years (RS-I)25. The
study was later extended to two stages and contained 14,926 subjects as of 2008.
Written informed consent was obtained from all participants, and the Medical
Ethics Committee of the Erasmus Medical Center, Rotterdam, approved the
study25. Metabolomics measurements were quantified in fasted EDTA plasma
samples using the Nightingale Health platform. We included all 2,949 samples with
complete baseline covariates and NMR metabolomics that were available in the
BBMRI-NL platform.
Leiden Longevity Study. The Leiden Longevity Study (LLS) consists of
421 long-lived families of European descent. Families were included if at least
two long-lived siblings were alive and fulfilled the age criterion of 89 years or
older for males and 91 years or older for females, representing <0.5% of the
Dutch population in 2001 (ref. 26). In total, 944 long-lived proband siblings
(mean age 94 years, range 89–104), 1,671 offspring (mean age 61 years, range
39–81) and 744 spouses thereof (mean age 60 years, range 36–79) were included.
Registry-based follow-up until 27 October 2016 was available for all participants.
Metabolites were successfully quantified in 843 nonagenarians, 1,157 of their
offspring and 684 controls using nonfasted EDTA plasma samples. We included
all 1,655 samples of the offspring and spouse population with complete baseline
covariates and NMR metabolomics available in the BBMRI-NL platform.
PROspective Study of Pravastatin in the Elderly at Risk. The PROspective
Study of Pravastatin in the Elderly at Risk (PROSPER) trial is a double-blind,
randomized, placebo-controlled trial investigating the benefit of pravastatin
(40 mg d–1) in elderly individuals at risk of CVD27,78. In total, 5,804 participants
(70–82 years) were identified in the primary care setting between December 1997
and May 1999 from three centers: Glasgow (UK) Cork (Ireland) and Leiden (the
Netherlands). The mean follow-up period was 3.2 years. All included patients
either had evidence of vascular disease (physician-diagnosed stable angina, stroke,
transient ischemic attack or myocardial infarction) or high risk of vascular disease
as determined by hypertension, diabetes or smoking status. Fasting venous blood
samples were collected at baseline and at 3-month intervals and stored at −80 °C.
For the present study, all individuals recruited at the Leiden recruitment center
and with NMR metabolomics data available through the BBMRI-NL consortium
(in total, 960 individuals) were included, employing the study as a cohort
study. NMR metabolomics was quantified from previously unthawed 6-month
postrandomization samples.
Reporting summary. Further information on research design is available in the
Nature Research Reporting Summary linked to this article.
Data availability
UK Biobank data, including NMR metabolomics, are publicly available to bona fide
researchers upon application at http://www.ukbiobank.ac.uk/using-the-resource/.
Detailed information on predictors and endpoints used in this study is presented
in Supplementary Tables 1–3. WHII data are available for the scientific
community, and researchers are invited to apply for data access at https://www.
dementiasplatform.uk/. Data from the BBMRI-NL consortium are available upon
application at https://www.bbmri.nl/Omics-metabolomics.
Code availability
All code developed and used throughout this study has been made open source and
is available on GitHub. The code used to train the metabolomic state model can
be found at github.com/thbuerg/MetabolomicsCommonDiseases, while the code
used to run analysis on trained models can be found at github.com/JakobSteinfeldt/
MetabolomicsCommonDiseases.
References
61. Moons, K. G. M. et al. Transparent Reporting of a multivariable prediction
model for Individual Prognosis Or Diagnosis (TRIPOD): explanation and
elaboration. Ann. Intern. Med. 162, W1–W73 (2015).
62. SCORE2 working group and ESC Cardiovascular risk collaboration. SCORE2
risk prediction algorithms: new models to estimate 10-year risk of
cardiovascular disease in Europe. Eur. Heart J. 42, 2439–2454 (2021).
63. Steinfeldt, J. et al. Neural network-based integration of polygenic and clinical
information: development and validation of a prediction model for 10-year
risk of major adverse cardiac events in the UK Biobank cohort. Lancet Digit
Health 4, e84–e94 (2022).
64. Stekhoven, D. J. & Bühlmann, P. MissForest–non-parametric missing value
imputation for mixed-type data. Bioinformatics 28, 112–118 (2012).
65. Srivastava, N., Hinton, G., Krizhevsky, A., Sutskever, I. & Salakhutdinov, R.
Dropout: a simple way to prevent neural networks from overtting. J. Mach.
Learn. Res. 15, 1929–1958 (2014).
66. Elfwing, S., Uchibe, E. & Doya, K. Sigmoid-weighted linear units for neural
network function approximation in reinforcement learning. ScienceDirect 108,
3–11 (2017).
67. He, K., Zhang, X., Ren, S. & Sun, J. Deep Residual Learning for Image
Recognition. In 2016 IEEE Conference on Computer Vision and Pattern
Recognition (CVPR) 770–778 (2016).
68. Katzman, J. L. et al. DeepSurv: personalized treatment recommender system
using a Cox proportional hazards deep neural network. BMC Med. Res.
Methodol. 18, 24 (2018).
69. Kingma, D. P. & Ba, J. L. Adam: a method for stochastic optimization.
Preprint at https://arxiv.org/abs/1412.6980v8 (2015).
70. Paszke, A. et al. Automatic dierentiation in PyTorch. Semanticscholar
https://www.semanticscholar.org/paper/Automatic-dierentiation-in-
PyTorch-Paszke-Gross/b36a5bb1707bb9c70025294b3a310138aae8327a
(2017).
71. Davidson-Pilon, C. et al. CamDavidsonPilon/lifelines: v0.25.8. (2021)
https://doi.org/10.5281/zenodo.4457577
72. R Core Team. R: A Language and Environment for Statistical Computing
(R Foundation for Statistical Computing, 2020).
73. Lundberg, S. M. et al. Explainable machine-learning predictions for the
prevention of hypoxaemia during surgery. Nat. Biomed. Eng. 2,
749–760 (2018).
74. Shrikumar, A., Greenside, P. & Kundaje, A. Learning important features
through propagating activation dierences. Preprint at https://arxiv.org/
abs/1704.02685. (2019).
75. Lundberg, S. shap: A game theoretic approach to explain the output of any
machine learning model. Github https://github.com/slundberg/shap (2022).
76. UMAP: Uniform Manifold Approximation and Projection for Dimension
Reduction—UMAP 0.5 documentation. UMAP https://umap-learn.readthedocs.
io/en/latest/index.html (2018).
77. onnx: Open standard for machine learning interoperability. Github
https://github.com/onnx/onnx (2022).
78. Shepherd, J. et al. e design of a prospective study of Pravastatin in the
Elderly at Risk (PROSPER). PROSPER Study Group. PROspective Study of
Pravastatin in the Elderly at Risk. Am. J. Cardiol. 84, 1192–1197 (1999).
Acknowledgements
This research was conducted using data from UK Biobank, a major biomedical
databas e (https://www.ukbiobank.ac.uk/) via application no. 51157. This project was
funded by Charité – Universitätsmedizin Berlin and Einstein Foundation Berlin. The
study was supported by the BMBF-funded Medical Informatics Initiative (HiGHmed,
nos. 01ZZ1802A–01ZZ1802Z). Resources of Flaticon.com were used in the design of
the figures. M. Kivimaki received relevant founding from The Wellcome Trust (no.
221854/Z/20/Z), the Medical Research Council (no. MR/R024227/1) and the US
National Institute on Aging (no. R01AG056477). E.B.v.d.A. received relevant funds from
the Dutch Research Council (NWO, no. VENI: 09150161810095). The WHII study was
supported by the Wellcome Trust (no. 221854/Z/20/Z), the Medical Research Council
(no. MR/R024227/1) and the US National Institute on Aging (no. R01AG056477).
Ethical approval for the WHII study was obtained from the University College London
Hospital Committee on the Ethics of Human Research and the NHS Health Research
Authority, London-Harrow Research Ethics Committee (nos. REC 85/0938 and IRAS
142374). Written informed consent from participants was obtained at each contact. The
authors thank A. Ryan and M. Newbury from the DPUK platform team for their swift
and immediate help with data access. The RS is supported by the Erasmus MC University
Medical Center and Erasmus University Rotterdam; the Netherlands Organization for
Scientific Research (NWO); the Netherlands Organization for Health Research and
Development (ZonMw); the Research Institute for Diseases in the Elderly (RIDE); the
Netherlands Genomics Initiative (NGI); the Ministry of Education, Culture and Science;
the Ministry of Health, Welfare and Sports; the European Commission (DG XII); and the
Municipality of Rotterdam. Metabolomics measurements were funded by Biobanking
and Biomolecular Resources Research Infrastructure (BBMRI)-NL (no. 184.021.007) and
JNPD under the project PERADES (grant no. 733051021, Defining Genetic, Polygenic
and Environmental Risk for Alzheimer’s Disease using multiple powerful cohorts,
focused Epigenetics and Stem cell metabolomics). The RS protocol was approved by the
Medical Ethics Committee of the Erasmus MC Rotterdam, the Netherlands (no. MEC
02.1015) and by the Dutch Ministry of Health, Welfare and Sport (Population Screening
Act WBO, license no. 1071272-159521-PG). In accordance with the Declaration of
Helsinki, the RS obtained written informed consent from all participants before their
entering the study. The LLS received funding from the European Union’s Seventh
Framework Program (FP7/2007-2011) under grant agreement no. 259679. This study
was supported by a grant from the Innovation-Oriented Research Program on Genomics
(SenterNovem, no. IGE05007), the Center for Medical Systems Biology and the
Netherlands Consortium for Healthy Ageing (grant nos. 05040202 and 050-060-810), all
within the framework of the Netherlands Genomics Initiative, NWO, Unilever Colworth
NATURE MEDICINE | www.nature.com/naturemedicine
Content courtesy of Springer Nature, terms of use apply. Rights reserved
Articles
Nature MediciNe
and by BBMRI-NL, a Research Infrastructure financed by the Dutch government (NWO,
no. 184.021.007). The LLS protocol was approved by the Medical Ethical Committee
of Leiden University Medical Center before the start of the study (no. P01.113). In
accordance with the Declaration of Helsinki, the LLS obtained informed consent from
all participants before their entry into the study. The PROSPER study was supported
by an investigator-initiated grant obtained from Bristol-Myers Squibb. J.W. Jukema is
an Established Clinical Investigator of the Netherlands Heart Foundation (grant no.
2001 D 032). PROSPER was supported by the European Federation of Pharmaceutical
Industries Associations (EFPIA), Innovative Medicines Initiative Joint undertaking,
European Medical Information Framework (EMIF, grant no. 115372) and the European
Commission under the Health Cooperation Work Program of the 7th Framework
Program (grant no. 305507) ‘Heart ‘omics’ in AGEing’ (HOMAGE). The PROSPER
protocol was approved by institutional ethics review boards of Cork University (Ireland),
Glasgow University (UK) and Leiden University Medical Center (the Netherlands). In
accordance with the Declaration of Helsinki, the PROSPER study obtained informed
consent from all participants before their entry into the study.
Author contributions
R.E., U.L., J.D., T.B. and J. Steinfeldt conceived and designed the project. T.B. and
J. Steinfeldt implemented models, conducted experiments and performed data analysis.
G.R. supported model introspection. M.P. advised analysis and helped in interpretation
of attribution profiles. D.B., D.V., S.T., S.P.M., N.S., J.W.J., B.L., M. Kavrusi, M.G., M.A.I.,
E.B.v.d.A. and E.S. helped with replication of analysis in the BBMRI-NL cohorts. J.U.z.B.,
L.L., N.H., P.K., L.C., H.S., J.M.B. and B.W. supported the analysis. M.P., J. Spranger, F.K.,
M. Kivimaki and C.L. provided methodological support and contributed to discussion
of the results. T.B., J. Steinfeldt, R.E. and U.L. wrote and prepared the manuscript. All
authors read, revised and approved the manuscript.
Competing interests
U.L. received grants from Bayer, Novartis and Amgen, consulting fees from Bayer, Sanofi,
Amgen, Novartis and Daichy Sankyo and honoraria from Novartis, Sanofi, Bayer, Amgen
and Daichy Sankyo. J.D. received consulting fees from GENinCode UK Ltd, honoraria
from Amgen, Boehringer Ingelheim, Merck, Pfizer, Aegerion, Novartis, Sanofi, Takeda,
Novo Nordisk and Bayer and is Chief Medical Advisor to Our Future Health. R.E.
received honoraria from Sanofi and consulting fees from Boehringer Ingelheim. All other
authors declare no competing interests.
Additional information
Extended data is available for this paper at https://doi.org/10.1038/s41591-022-01980-3.
Supplementary information The online version contains supplementary material
available at https://doi.org/10.1038/s41591-022-01980-3.
Correspondence and requests for materials should be addressed to Roland Eils.
Peer review information Nature Medicine thanks Jessica Lasky-Su and the other,
anonymous, reviewer(s) for their contribution to the peer review of this work. Primary
Handling Editor: Michael Basson, in collaboration with the Nature Medicine team.
Reprints and permissions information is available at www.nature.com/reprints.
NATURE MEDICINE | www.nature.com/naturemedicine
Content courtesy of Springer Nature, terms of use apply. Rights reserved
Articles Nature MediciNe
Extended Data Fig. 1 | Details of the metabolomic state model. a) Overview of the residual architecture of the metabolomic state model. 168 circulating
metabolomic markers are fed to the shared trunk network to learn a common shared representation. Endpoint-specific head networks then predict the
metabolomic state for each endpoint from the shared representation and the input using a residual connection. b) Details of the residual head network.
The model architecture is described in detail in (Methods Section ‘Metabolomic state model’).
NATURE MEDICINE | www.nature.com/naturemedicine
Content courtesy of Springer Nature, terms of use apply. Rights reserved
Articles
Nature MediciNe
Extended Data Fig. 2 | The metabolomic state model outperforms linear baselines on NMR-derived metabolite profiles, and NMR-derived metabolite
profiles are more predictive than PANEL metabolites. a) Displayed are C-indices for the Cox Proportional Hazards models trained on the metabolomic
state (MET), the 168 metabolites (CPH) as well as on the first ten components of a PCA-reduction of the 168 metabolites (PCA) for each of the 24
investigated endpoints. The metabolomic state performs comparably or better than both the CPH and PCA models for all endpoints, except prostate
cancer. b) Displayed are C-indices for Cox Proportional Hazards models trained on Age+Sex (Age+Sex), the metabolomic states derived from NMR
metabolomics (MET(NMR)), the metabolomic states derived from the PANEL metabolites (MET(PANEL)) and combinations of Age+Sex and the
metabolomic states respectively. NMR profiles provide predictive information comparable or superior to the PANEL metabolites for all investigated
endpoints, also reflected in the predictive performance over the Age+Sex covariates. The MET(PANEL) set included albumin, cholesterol, HDL and
LDL cholesterol, triglycerides, glucose, and creatinine. Statistical measures were derived from n = 117.981 individuals. Individuals with prior events were
excluded (Supplementary Table 1). Data are presented as median (center of error bar) and 95% CI (line of error bar) determined by bootstrapping over
1000 iterations. PAD - Peripheral Artery Disease, AAA - Abdominal Aortic Aneurysm, COPD - Chronic Obstructive Pulmonary Disease.
NATURE MEDICINE | www.nature.com/naturemedicine
Content courtesy of Springer Nature, terms of use apply. Rights reserved
Articles Nature MediciNe
Extended Data Fig. 3 | External validation in four independent cohorts. a) Displayed are discriminative performances described by the C-index for
UK Biobank and the four external validation cohorts, Whitehall II (WHII), Rotterdam Study (RS), Leiden Longevity Study (LLS), and the PROSPER trial
(PROSPER). CPH models were trained on the metabolomic state model (MET) as fitted on UK Biobank and applied to each cohort, as well as on Age+Sex
and Age+Sex+MET. The metabolomic state is predictive in the replication cohorts for all assessed endpoints. Dots indicate the median performance, while
whiskers indicate the 95% confidence interval (CI) determined by bootstrapping over 1000 iterations. b) Age+Sex adjusted hazard ratios (HRs) for the
metabolomic state in all five cohorts. A unit standard deviation increase in the metabolomic state corresponds to an HR increase in predicted risk. Statistical
measures were derived from n = 6.117 (Whitehall II), n = 2949 (Rotterdam Study), n = 1655 (Leiden Longevity Study), and n = 960 (PROSPER) individuals as
indicated. Data are presented as median (center of error bar) and 95% CI (line of error bar) determined by bootstrapping over 1000 iterations.
NATURE MEDICINE | www.nature.com/naturemedicine
Content courtesy of Springer Nature, terms of use apply. Rights reserved
Articles
Nature MediciNe
Extended Data Fig. 4 | See next page for caption.
NATURE MEDICINE | www.nature.com/naturemedicine
Content courtesy of Springer Nature, terms of use apply. Rights reserved
Articles Nature MediciNe
Extended Data Fig. 4 | The discriminative performance is largely comparable over multiple subgroups. Discriminative performance is stratified by
endpoint, age at recruitment, biological sex, and self-reported ethnic background. As the concordance index is only reliable if a sufficient number of events
are recorded, subgroups with < 100 events were excluded. The number of events and eligible individuals is indicated at the top of each panel. Data are
presented as median (center of error bar) and 95% CI (line of error bar) determined by bootstrapping with 1000 iterations.
NATURE MEDICINE | www.nature.com/naturemedicine
Content courtesy of Springer Nature, terms of use apply. Rights reserved
Articles
Nature MediciNe
Extended Data Fig. 5 | Comparison of the predictive performance of the PANEL predictors in a Cox proportional hazard model and the neural
network. Comparison of discriminative performances of the CPH models and Metabolomic State Model (MSM) trained on the PANEL covariates.
The discriminative performance of the PANEL predictors is either similar or can be further improved by modeling with the same architecture as the
metabolomic state model for most (non-cancer) endpoints. Statistical measures were derived from n = 117.981 individuals. Individuals with prior events
were excluded (Supplementary Table 1). Data are presented as median (center of error bar) and 95% CI (line of error bar) determined by bootstrapping
with 1000 iterations.
NATURE MEDICINE | www.nature.com/naturemedicine
Content courtesy of Springer Nature, terms of use apply. Rights reserved
Articles Nature MediciNe
Extended Data Fig. 6 | See next page for caption.
NATURE MEDICINE | www.nature.com/naturemedicine
Content courtesy of Springer Nature, terms of use apply. Rights reserved
Articles
Nature MediciNe
Extended Data Fig. 6 | Adjusted effect of the metabolomic state is endpoint dependent. a) Adjusted trajectories representing the partial cumulative risk
dependent on the metabolomic state over time for the endpoints where the metabolomic state added information to the Age+Sex baseline (see Fig. 3b)
for the bottom (light blue), median (blue), and top (dark blue) 10% metabolomic state quantiles. The shaded area indicates the 95% confidence interval
as estimated by bootstrapping over 1000 iterations. b) Adjusted hazard ratios (HRs) for the metabolomic state in combination with the three clinical
predictor sets. A unit standard deviation increase in the metabolomic state corresponds to an HR increase in predicted risk. Statistical measures were
derived from n = 117.981 individuals. Individuals with prior events were excluded (Supplementary Table 1). Data are presented as median (center of error
bar) and 95% CI (line of error bar) determined by bootstrapping with 1000 iterations.
NATURE MEDICINE | www.nature.com/naturemedicine
Content courtesy of Springer Nature, terms of use apply. Rights reserved
Articles Nature MediciNe
-0.02 0.00 0.02
∆ C-Index
PANEL PANEL+APOE4
5 10 5 10
0.5
1
1.5
2
Time [Years]
Adjusted Partial Trajectories [%]
0.5 1.0 1.5 2.0
Adjusted Hazard Ratio / SD(Metabolomic State)
0
25
50
75
100
0 2.5 5 7.5 10
Threshold Probability [%]
Standardized Net Benefit [%]
b
a
c
d
PANEL+APOE4
PANEL+APOE4+MET
Net Benefit of MET
addition to PANEL
Net Benefit of MET
addition to PANEL+APOE4
Median 10% Bottom 10%
Metabolomic State Quantile
Top 10%
PANEL
PANEL+MET
PANEL+APOE4
PANEL+APOE4+MET
HRPANEL+MET: 1.46 (1.43, 1.58)
HRPANEL+APOE4+MET: 1.43 (1.41, 1.54)
PANEL
PANEL+MET
Extended Data Fig. 7 | The metabolomic state contains independent predictive information over the APOE4 carrier status for all-cause
dementia. a) Displayed are C-index deltas between the CPH model trained on the PANEL + APOE4 predictor set, its metabolomic state addition
(PANEL + APOE4 + MET), and CPH models trained on the PANEL set and its respective metabolomic state addition (PANEL + MET). The metabolomic
state adds predictive information over the PANEL + APOE4. b) Partial trajectory for MET deciles (Top, Median, Bottom 10%) adjusted for PANEL and
PANEL + APOE4, respectively. c) Hazard Ratio for the Metabolomic State adjusted for the predictors of the PANEL and PANEL + APOE4. d) Decision
curve analysis for PANEL/PANEL + MET and PANEL + APOE4/PANEL + APOE4 + MET. The areas in between the solid and dotted lines indicate added net
benefits resulting from metabolomic state addition to PANEL (gray lines, red area) and PANEL + APOE4 (black lines, violet area), respectively. Adding
MET to PANEL improves net population benefit between the 2–8% decision threshold. In the case of PANEL + APOE4, MET addition improves utility at
thresholds between 5–10%. Statistical measures were derived from n = 117.245 individuals without dementia at recruitment. Data are presented as median
(center of error bar) and 95% CI (line of error bar) determined by bootstrapping with 1000 iterations.
NATURE MEDICINE | www.nature.com/naturemedicine
Content courtesy of Springer Nature, terms of use apply. Rights reserved
Articles
Nature MediciNe
Extended Data Fig. 8 | See next page for caption.
NATURE MEDICINE | www.nature.com/naturemedicine
Content courtesy of Springer Nature, terms of use apply. Rights reserved
Articles Nature MediciNe
Extended Data Fig. 8 | Global metabolite importances for each metabolite and endpoint. Heatmap of the metabolite importances, represented by
absolute global SHAP value estimates per endpoint for the 168 circulating metabolites. The endpoints are sorted by the discriminative performance of the
metabolomic state (left to right, see Fig. 3a). MACE - Major Adverse Cardiac Events, CHD - Coronary Heart Disease, PAD - Peripheral Artery Disease,
AAA - Abdominal Aortic Aneurysm, COPD - Chronic Obstructive Pulmonary Disease.
NATURE MEDICINE | www.nature.com/naturemedicine
Content courtesy of Springer Nature, terms of use apply. Rights reserved
Articles
Nature MediciNe
Extended Data Fig. 9 | See next page for caption.
NATURE MEDICINE | www.nature.com/naturemedicine
Content courtesy of Springer Nature, terms of use apply. Rights reserved
Articles Nature MediciNe
Extended Data Fig. 9 | Individual attribution profiles diverge for high-risk individuals in T2D. The UMAP projection allows an assessment of the
complex, high-dimensional manifold of attribution values in 2-dimensional space. For visualization, 41 unconnected outliers of 117981 total observations
were excluded. a) UMAP of the SHAP value metabolite attributions for T2D for the entire study population colored by each individual’s metabolomic
state. b) The same UMAP colored by the Glucose SHAP value. c) Displays individual attribution profiles for three high-risk (metabolomic state > 10, top
1% metabolomic state percentile) individuals, indicated by the letters A, B, C in the central UMAP. The three individual attribution profiles are dominated
by different metabolites. The scale bar represents a unit in the UMAP space. The individual attribution profiles are set up equivalently to Figure 6: Each
point in an individual attribution profile indicates one metabolite; the position, size, and color of the point indicate the magnitude and direction of the
attributed contribution to predicted risk. The green and red circles represent the bounds of the top and bottom percentile of the global SHAP distribution,
respectively, indicating outliers in the SHAP global distribution.
NATURE MEDICINE | www.nature.com/naturemedicine
Content courtesy of Springer Nature, terms of use apply. Rights reserved
Articles
Nature MediciNe
Extended Data Fig. 10 | See next page for caption.
NATURE MEDICINE | www.nature.com/naturemedicine
Content courtesy of Springer Nature, terms of use apply. Rights reserved
Articles Nature MediciNe
Extended Data Fig. 10 | Metabolites differ throughout the attribution space. Displayed are distributions for all measured metabolites (n = 168) stratified
by the region (A, B, and C) in the attribution space, defined by the UMAP of the attributions for T2D (see Extended Data Figure 9c). Regions were defined
by including all samples with an euclidean distance < 1 to the centroid A, B, and C, respectively; a Euclidean distance of 1 is indicated by the scale bar
(see Extended Data Figure 9c). The distributions differ notably for metabolites, including glucose, fatty acids (that is LA and Omega-6), and multiple
lipoprotein components (that is VLDL cholesterol and very large HDL triglycerides).
NATURE MEDICINE | www.nature.com/naturemedicine
Content courtesy of Springer Nature, terms of use apply. Rights reserved
Content courtesy of Springer Nature, terms of use apply. Rights reserved
Content courtesy of Springer Nature, terms of use apply. Rights reserved
1.
2.
3.
4.
5.
6.
Terms and Conditions
Springer Nature journal content, brought to you courtesy of Springer Nature Customer Service Center GmbH (“Springer Nature”).
Springer Nature supports a reasonable amount of sharing of research papers by authors, subscribers and authorised users (“Users”), for small-scale
personal, non-commercial use provided that all copyright, trade and service marks and other proprietary notices are maintained. By accessing,
sharing, receiving or otherwise using the Springer Nature journal content you agree to these terms of use (“Terms”). For these purposes, Springer
Nature considers academic use (by researchers and students) to be non-commercial.
These Terms are supplementary and will apply in addition to any applicable website terms and conditions, a relevant site licence or a personal
subscription. These Terms will prevail over any conflict or ambiguity with regards to the relevant terms, a site licence or a personal subscription (to
the extent of the conflict or ambiguity only). For Creative Commons-licensed articles, the terms of the Creative Commons license used will apply.
We collect and use personal data to provide access to the Springer Nature journal content. We may also use these personal data internally within
ResearchGate and Springer Nature and as agreed share it, in an anonymised way, for purposes of tracking, analysis and reporting. We will not
otherwise disclose your personal data outside the ResearchGate or the Springer Nature group of companies unless we have your permission as
detailed in the Privacy Policy.
While Users may use the Springer Nature journal content for small scale, personal non-commercial use, it is important to note that Users may not:
use such content for the purpose of providing other users with access on a regular or large scale basis or as a means to circumvent access control;
use such content where to do so would be considered a criminal or statutory offence in any jurisdiction, or gives rise to civil liability, or is
otherwise unlawful;
falsely or misleadingly imply or suggest endorsement, approval , sponsorship, or association unless explicitly agreed to by Springer Nature in
writing;
use bots or other automated methods to access the content or redirect messages
override any security feature or exclusionary protocol; or
share the content in order to create substitute for Springer Nature products or services or a systematic database of Springer Nature journal content.
In line with the restriction against commercial use, Springer Nature does not permit the creation of a product or service that creates revenue, royalties,
rent or income from our content or its inclusion as part of a paid for service or for other commercial gain. Springer Nature journal content cannot be
used for inter-library loans and librarians may not upload Springer Nature journal content on a large scale into their, or any other, institutional
repository.
These terms of use are reviewed regularly and may be amended at any time. Springer Nature is not obligated to publish any information or content on
this website and may remove it or features or functionality at our sole discretion, at any time with or without notice. Springer Nature may revoke this
licence to you at any time and remove access to any copies of the Springer Nature journal content which have been saved.
To the fullest extent permitted by law, Springer Nature makes no warranties, representations or guarantees to Users, either express or implied with
respect to the Springer nature journal content and all parties disclaim and waive any implied warranties or warranties imposed by law, including
merchantability or fitness for any particular purpose.
Please note that these rights do not automatically extend to content, data or other material published by Springer Nature that may be licensed from
third parties.
If you would like to use or distribute our Springer Nature journal content to a wider audience or on a regular basis or in any other manner not
expressly permitted by these Terms, please contact Springer Nature at
onlineservice@springernature.com
