Genome-Wide Polygenic Score for Muscle Strength Predicts Risk for Common Diseases and Lifespan: A Prospective Cohort Study

Abstract Background We used a polygenic score for hand grip strength (PGS HGS) to investigate whether genetic predisposition for higher muscle strength predicts age-related noncommunicable diseases, survival from acute adverse health events, and mortality. Methods This study consisted of 342 443 Finnish biobank participants from FinnGen Data Freeze 10 (53% women) aged 40–108 with combined genotype and health registry data. Associations between PGS HGS and a total of 27 clinical endpoints were explored with linear or Cox regression models. Results A higher PGS HGS was associated with a reduced risk of selected common noncommunicable diseases and mortality by 2%–10%. The risk for these medical conditions decreased by 5%–23% for participants in the highest PGS HGS quintile compared to those in the lowest PGS HGS quintile. A 1 standard deviation (SD) increase in the PGS HGS predicted a lower body mass index (β = −0.112 kg/m2, standard error [SE] = 0.017, p = 1.69E-11) in women but not in men (β = 0.004 kg/m2, p = .768). PGS HGS was not associated with better survival after acute adverse health events compared to the nondiseased period. Conclusions The genotype that supports higher muscle strength appears to protect against future health adversities, albeit with modest effect sizes. Further research is needed to investigate whether or how a favorable lifestyle modifies this intrinsic capacity to resist diseases, and if the impacts of lifestyle behavior on health differs due to genetic predisposition for muscle strength.


FinnGen Endpoint Definitions
Below are the FinnGen endpoint definitions for selected metabolic, cardiovascular, pulmonary as well as musculoskeletal and connective tissue diseases and cancers, which were used in our analysis.Endpoints are based on the Hospital Discharge registry, the Cause of Death registry, Cancer registry and the Social Insurance Institution of Finland (KELA) registry for reimbursements of medical expenses.More detailed information about the FinnGen endpoints definitions and codes can be found in the FinnGen webpages https://risteys.finngen.fi/ .• Definition: type II diabetes mellitus: A type of diabetes mellitus that is characterized by insulin resistance or desensitization and increased blood glucose levels.This is a chronic disease that can develop gradually over the life of a patient and can be linked to both environmental factors and heredity.

Ischemic heart diseases
• Definition: coronary thrombosis: Coagulation of blood in any of the coronary vessels.The presence of a blood clot (thrombus) often leads to myocardial infarction.

Hypertension
• Definition: Persistently high systemic arterial blood pressure.Based on multiple readings (blood pressure determination), hypertension is currently defined as when systolic pressure is consistently greater than 140 mm Hg or when diastolic pressure is consistently 90 mm Hg or more.

COPD
• Definition: chronic obstructive pulmonary disease: A chronic and progressive lung disorder characterized by the loss of elasticity of the bronchial tree and the air sacs, destruction of the air sacs wall, thickening of the bronchial wall, and mucous accumulation in the bronchial tree.The pathologic changes result in the disruption of the air flow in the bronchial airways.Signs and symptoms include shortness of breath, wheezing, productive cough, and chest tightness.The two main types of chronic obstructive pulmonary disease are chronic obstructive bronchitis and emphysema.

Asthma
• Definition: A bronchial disease that is characterized by chronic inflammation and narrowing of the airways, which is caused by a combination of environmental and genetic factors resulting in recurring periods of wheezing (a whistling sound while breathing), chest tightness, shortness of breath, mucus production and coughing.The symptoms appear due to a variety of triggers such as allergens, irritants, respiratory infections, weather changes, exercise, stress, reflux disease, medications, foods and emotional anxiety.

Arthrosis
• Definition: Hemarthrosis: Bleeding into the joints.It may arise from trauma or spontaneously in patients with hemophilia.

Rheumatoid arthritis
• Definition: rheumatoid arthritis: A chronic, systemic autoimmune disorder characterized by inflammation in the synovial membranes and articular surfaces.It manifests primarily as a symmetric, erosive polyarthritis that spares the axial skeleton and is typically associated with the presence in the serum of rheumatoid factor.

Osteoporosis
• Definition: A condition of reduced bone mass, with decreased cortical thickness and a decrease in the number and size of the trabeculae of cancellous bone (but normal chemical composition), resulting in increased fracture incidence.Osteoporosis is classified as primary (Type 1, postmenopausal osteoporosis; Type 2, age-associated osteoporosis; and idiopathic, which can affect juveniles, premenopausal women, and middle-aged men) and secondary osteoporosis (which results from an identifiable cause of bone mass loss).

Depression
• Definition: unipolar depression: A mood disorder having a clinical course involving one or more episodes of serious psychological depression that last two or more weeks each, do not have intervening episodes of mania or hypomania, and are characterized by a loss of interest or pleasure in almost all activities and by some or all of disturbances of appetite, sleep, or psychomotor functioning, a decrease in energy, difficulties in thinking or making decisions, loss of self-esteem or feelings of guilt, and suicidal thoughts or attempts.

Alzheimer's disease
• Definition: A progressive, neurodegenerative disease characterized by loss of function and death of nerve cells in several areas of the brain leading to loss of cognitive function such as memory and language.

Malignant neoplasm of bronchus and lung (controls excluding all cancers)
• Definition: No definition available.

Ethical Permits of the FinnGen Study
Patients and control subjects in FinnGen and FINRISK provided informed consent for biobank research, based on the Finnish Biobank Act.Alternatively, separate research cohorts, collected prior the Finnish Biobank Act came into effect (in September 2013) and start of FinnGen (August 2017), were collected based on studyspecific consents and later transferred to the Finnish biobanks after approval by Fimea (Finnish Medicines Agency), the National Supervisory Authority for Welfare and Health.Recruitment protocols followed the biobank protocols approved by Fimea.

Genotyping and Quality Control of the FinnGen Data
Chip genotyping was done using several Illumina and Affymetrix FinnGen Axiom arrays.The algorithms for genotype calling were GenCall or GenCall+zCall for Illumina and AxiomGT1 for Affymetrix chip genotypes.
The genome build of all genotypes was set to GRCh38/hg38.Quality control exclusions were done samplewise: samples with a call rate below 95% and heterozygosity test method-of-moments F coefficient estimate value deviated more than ±4SD from the mean were removed along with the samples which failed sex check or were among the multi-dimensional scaling principal component analysis outliers, and in variant-wise: variants with a call rate below 98%, minor allele count below 3 and Hardy-Weinberg Equilibrium p-value lower than 1e-06 were removed.
Pre-phasing was performed using Eagle v2.3.5 and imputation with Beagle v4.1 (protocol described in dx.doi.org/10.17504/protocols.io.xbgfijw) using Sisu v4 as reference panel which consists of 8,554 Finnish whole genome sequences (depth up to 30x).As a post-imputation quality control, variants with imputation quality score below 0.7 were removed.

Illness-Death Model and Results of the Time-Dependent Survival Analysis eFigure 1.
An extension of the illness-death model used in the analysis.
eTable 1 shows the characteristics of the participants according to diseased state (non-diseased, survived the first post acute event year, and died during the first post acute event year).In total, 9 992 (14.8%) participants who faced acute ischemic heart disease and 5 877 (14.4%) participants with stroke died during the first year after the event.Of the participants with femur fracture, 2 007 (30.8%) died during the first year post-fracture.In all participants, those who died during the first year after an acute event were significantly older and were likely to be former or current smokers compared to the first-year post acute event survivors and non-diseased participants.PGS HGS as a predictor of noncommunicable diseases and conditions and mortality in the FinnGen cohort.
Multivariable Cox regression analysis.The start of follow-up from age at baseline data collection, which is also the blood sampling for DNA analysis.Adjusted for sex, year of DNA sample collection, genotyping batch, and ten genetic principal components of ancestry.HR=Hazard Ratio, CI=Confidence Interval, q-value = adjusted p-value for the False Discovery Rate (FDR <0.05).PGS HGS as a predictor of noncommunicable diseases and conditions and mortality in the FINRISK cohort.

Results of the
Multivariable Cox regression analysis.The start of follow-up from birth.Adjusted for sex, year of DNA sample collection, genotyping batch, and ten genetic principal components of ancestry.HR=Hazard Ratio, CI=Confidence Interval, q-value = adjusted p-value for the False Discovery Rate (FDR <0.05).
PGS HGS as a predictor of noncommunicable diseases and conditions and mortality in the FINRISK cohort.
Multivariable Cox regression analysis.The start of follow-up from age at baseline data collection, which is also the blood sampling for DNA analysis.Adjusted for sex, year of DNA sample collection, genotyping batch, and ten genetic principal components of ancestry.HR=Hazard Ratio, CI=Confidence Interval, q-value = adjusted p-value for the False Discovery Rate (FDR <0.05.

Additional Analysis: Description and Results of the Association Analysis Between PGS BMI and HGS for Bi-Directional Associations
We investigated associations between measured BMI and HGS as well as PGS BMI and HGS in the Finnish Twin Study on Aging (FITSA) (1) cohort among 429 Finnish women, aged from 63 to 76 years (eTable 5) to be able to observe any bi-directional association.Polygenic scoring was conducted using GWAS meta-analysis summary statistics for BMI from the GIANT consortium website (2) with SBayesR approach (3).The total number of genetic variants used in the PGS BMI calculation was 902 833.A detailed description of genotyping, quality control, and polygenic scoring is presented elsewhere (4,5).Association analysis was performed using linear mixed modelling including the family number in the models as a random factor.The models were adjusted for age and also for ten genetic principal components, when PGS was included in the model.PGS value was standardized (z-score) and the level of significance was set at P<0.05.In the FITSA cohort, BMI did not correlate with measured HGS (β -0.369, SE 0.553, P=0.506).We did not find an association between PGS HGS and BMI (β 0.076, SE 0.242, P=0.753), and neither between PGS BMI and HGS (β -0.746, SE 3.142, P=0.812).

•
Definition: A disorder involving an excessive amount of body fat • FinnGen code: E4_Obesity • Hospital Discharge registry & Cause of Death registry:

307 •
Hospital Discharge registry & Cause of Death registry: o ICD-10: G30 o ICD-9: 3310 Dementia • Definition: obsolete_dementia: ['An acquired organic mental disorder with loss of intellectual abilities of sufficient severity to interfere with social or occupational functioning.The dysfunction is multifaceted and involves memory, behavior, personality, judgment, attention, spatial relations, language, abstract thought, and other executive functions.The intellectual decline is usually progressive, and initially spares the level of consciousness.']• FinnGen code: F5_DEMENTIA • Hospital Discharge registry & Cause of Death registry: o ICD-10: F00-F09 o ICD-9: 290|3310|4378A o ICD-8: 290 • KELA reimbursements: Kela codes o Medicine purchases : ATC o N06D Vascular dementia • Definition: A degenerative vascular disorder affecting the brain.It is caused by the blockage of the blood supply to the brain.It is manifested with decline of memory and cognitive functions.• FinnGen code: I9_VASCDEM • Hospital Discharge registry & Cause of Death registry: o ICD-10: F01 Cancers Colorectal cancer (controls excluding all cancers)

eTable 1. Characteristics of FinnGen Participants Stratified into Those Who Did Not Sustain an Acute Event (Non-Diseased), Those Who Sustained an Acute Event and Either Survived the first Post Acute Year (Diseased, Survived) or Died During the First Post Acute Event Year (Diseased, Died During the First Year)
Note: *One-way analysis of variance, § Welch test for variables with unequal variances between groups, ⸸ Chi square test, PGS HGS = Polygenic Scores for Hand Grip Strength, SD=Standard Deviation, BMI= Body Mass Index.

22 (2.13-5.87) <0.001 Sex Women Men 1.98 (1.93-2.02) <0.001
Extended Cox regression analysis.Reference group: sex, women; non-diseased state (no acute adverse health event during the follow-up and time before an acute event occurrence of the participants sustaining an acute event).Adjusted for sex, year of DNA sample collection, genotyping batch, and ten genetic principal components of ancestry.PGS HGS = Polygenic Scores for Hand Grip Strength, HR=Hazard Ratio, CI=Confidence Interval.Statistically significant values are shown in bold. Note:6.

Cumulative Incidence Curves from the Main Analysis and Results of the Sensitivity Analysis in the FinnGen Study eFigure 2.
Cumulative incidence of diseases and mortality by PGS HGS categories in FinnGen for the scaled age.Cumulative incidence is presented as a percentage.The survival curves are from the multivariable Cox regression analysis.The start of follow-up from birth.Only endpoints, which provided statistical significance are presented.Adjusted for sex, year of DNA sample collection, genotyping batch, and ten genetic principal components of ancestry.

eTable 3. Characteristics of Participants in the FinnGen Study When Start of the Follow-Up Was Set to the Age at the Blood Sampling for DNA Analysis
Age at the time of death or at the end of follow-up on 31 December 2021.Phenotype data were obtained from the biobanks https://www.finngen.fi/en/data_protection/data-protection-statement),SD=Standard Deviation, BMI=Body Mass Index. Note:

Sensitivity Analysis in the FINRISK Study eTable 4. Characteristics of Participants in the Population-Based FINRISK Study
Age at the time of death or at the end of follow-up on 31 December 2021.Phenotype data were obtained from the THL Biobank (https://thl.fi/en/web/thlfi-en/research-and-development/research-and-projects/the-national-finrisk-study),SD=Standard Deviation, BMI=Body Mass Index. Note: FITSA=Finnish Twin Study on Aging, SD=Standard Deviation, BMI= Body Mass Index, HGS= Hand Grip Strength. Note: