The high prevalence (>5%) of autoimmune hypothyroidism (AIHT) provides a unique opportunity to dissect genetic contributions to systemic and organ-specific autoimmunity. Here we performed a genome-wide association meta-analysis of 81,718 AIHT cases in FinnGen and the UK Biobank, identifying 418 independent signals (P < 5 × 10-8). At 48 of these loci, a protein-coding variant is, or is highly correlated (r2 > 0.95) with, the lead variant, including Finnish-enriched coding variants in LAG3, ZAP70 and TG. We demonstrated that ZAP70:T155M reduces T cell activation and broadly compare large-scale scans of nonthyroid autoimmunity and thyroid-stimulating hormone levels with a Bayesian classifier to assign loci into distinct groupings, estimating that 38% are involved in general autoimmunity whereas 20% are thyroid specific. We further identified substantial antagonistic pleiotropy, with 10% of AIHT loci showing a consistent protective effect against skin cancer. The AIHT results, including numerous genes encoding checkpoint proteins, support the causal role of natural immune variation influencing cancer outcomes.

ABSTRACT Although the Y chromosome represents roughly 2% of the male genome, it is often ignored in genome-wide association studies (GWAS). Subsequently, the potential health impacts of Y-chromosomal genetic variation remain incompletely understood. To fill this gap, we performed a phenome-wide association study (PheWAS) in FinnGen across 1,426 binary and quantitative traits using Y-chromosomal variation (frequency ≥ 1%) in 104,334 genotyped men. As Y chromosome variation is prone to population stratification, we performed carefully adjusted association analyses and further examined these through kin-based validation in 19,275 female and 24,712 male 1 st degree relatives. We found 121 suggestive (p < 5.6x10 -3 ) phenotypic associations in the Y chromosome, yet none of these were strong enough to reach phenome-wide significance (p < 3.9x10 -6 ). While only 38 associations were supported in the kin-based validation, intriguingly we found support for a previously suggested link between haplogroup I1 and coronary heart disease (CHD; OR=1.06, 95%CI=1.02-1.11, p=3.7x10 -3 ; male validation OR=1.05; female validation OR=0.97). The I1-CHD association was detected across distinct geographical areas within Finland and was independent from Loss of Y (LOY) and the autosomal risk to CHD, proposing a link between germline Y-chromosomal variation and heart disease risk. Overall, this study presents a comprehensive phenome-wide analysis of Y-chromosomal associations, highlighting the potential relevance of Y-chromosomal variation beyond sex determination. Our findings further emphasize the need for improved capture of Y-chromosomal variants and further analyses in biobank-scale data to allow for deeper exploration of male-specific genetic architecture of complex diseases. GRAPHICAL ABSTRACT Graphical Abstract Overview of the study. Using a dataset of 104,334 unrelated genotyped Finnish men we performed a PheWAS across 1,426 binary and quantitative traits and common Y-chromosomal variation (MAF ≥ 1%), adjusting for geographical confounding factors known to vary within the population. To confirm suggestive associations identified in the PheWAS, we performed kin-based validation to infer whether the association is likely linked to the Y chromosome, using the samples’ 1st degree male (positive control) and 1st degree female (negative control) relatives’ phenotype data.

Migraine is a highly prevalent neurovascular disorder for which genome-wide association studies (GWAS) have identified over one hundred risk loci, yet the causal variants and genes remain mostly unknown. Here, we meta-analyze three migraine GWAS including 98,374 cases and 869,160 controls and identify 122 independent risk loci of which 35 were new. Fine-mapping of a meta-analysis is challenging because some variants may be missing from some participating studies and accurate linkage disequilibrium (LD) information of the variants is often not available. Here, using the exact in-sample LD, we first investigate which statistics could reliably capture the quality of fine-mapping when only reference LD is available. We observe that the posterior expected number of causal variants best distinguishes between the high- and low-quality results. Next, we perform fine-mapping for 102 autosomal risk regions using FINEMAP. We produce high-quality fine-mapping for 93 regions and define 181 distinct credible sets. Among the high-quality credible sets are 7 variants with very high posterior inclusion probability (PIP > 0.9) and 2 missense variants with PIP > 0.5 (rs6330 in NGF and rs1133400 in INPP5A). For 35 association signals, we manage to narrow down the set of potential risk variants to at most 5 variants. A large-scale migraine genetics study identifies 122 risk regions, including 35 previously unreported ones, and pinpoints likely causal genetic variants using an in silico fine-mapping method.

The X chromosome (chrX) is often excluded from genome-wide association studies due to its unique biology complicating the analysis and interpretation of genetic data. Consequently, the influence of chrX on human complex traits remains debated. Here, we systematically assessed the relevance of chrX and the effect of its biology on complex traits by analyzing 48 quantitative traits in 343,695 individuals in UK Biobank with replication in 412,181 individuals from FinnGen. We show that, in the general population, chrX contributes to complex trait heritability at a rate of 3% of the autosomal heritability, consistent with the amount of genetic variation observed in chrX. We find that a pronounced male bias in chrX heritability supports the presence of near-complete dosage compensation between sexes through X chromosome inactivation (XCI). However, we also find subtle yet plausible evidence of escape from XCI contributing to human height. Assuming full XCI, the observed chrX contribution to complex trait heritability in both sexes is greater than expected given the presence of only a single active copy of chrX, mirroring potential dosage compensation between chrX and the autosomes. We find this enhanced contribution attributable to systematically larger active allele effects from chrX compared to autosomes in both sexes, independent of allele frequency and variant deleteriousness. Together, these findings support a model in which the two dosage-compensation mechanisms work in concert to balance the influence of chrX across the population while preserving sex-specific differences at a manageable level. Overall, our study advocates for more comprehensive locus discovery efforts in chrX.

Motivation The most informative genome-wide association studies (GWAS) are meta-analyses that have combined multiple studies to increase the GWAS sample size. Statistical fine-mapping is a key downstream analysis of GWAS to jointly evaluate the probability of causality of all variants in a genomic region of interest. Current fine-mapping methods are miscalibrated in the meta-analysis setting due to variation in sample size across the variants.Results We introduce FINEMAP-miss, a new fine-mapping method that extends the FINEMAP model to account for variant-specific missingness. We show that FINEMAP-miss is well-calibrated in meta-analysis simulations where the standard fine-mapping fails. Compared to the summary statistics imputation approach, FINEMAP-miss provides clear improvement when the causal variants have low imputation information or when the sample size or complexity of the meta-analysis setting increase. We successfully apply FINEMAP-miss on a breast cancer GWAS meta-analysis where neither the standard fine-mapping nor the summary statistics imputation are applicable.Availability An open source implementation of FINEMAP-miss as an R package ("finemapmiss") is available at https://github.com/JoonasKartau/finemapmiss. The archived version of FINEMAP-miss used for this publication can be found on Zenodo at https://doi.org/10.5281/zenodo.17492622.Supplementary Information is available at the journal's web site.

BACKGROUND:Genome-wide association studies of lipid species have identified several loci shared with various diseases, however, the relationship between lipid species and disease risk remains poorly understood. Here we investigated whether the plasma levels of lipid species are causally linked to disease risk. METHODS:We built genetic predictors of 179 lipid species, measured in 7174 Finnish individuals, by utilising either 11 high-impact genomic loci or genome-wide polygenic scores (PGS). We assessed the impact of the lipid species on seven diseases by performing disease association across FinnGen (n = 500,348), UK Biobank (n = 420,531), and Generation Scotland (n = 20,032). We performed univariable Mendelian randomisation (MR) and multivariable MR (MVMR) analyses to examine whether lipid species impact disease risk independently of standard lipids. FINDINGS:PGS explained >4% of the variance for 34 lipid species but variants outside the high-impact loci had only a marginal contribution. Variants within the high-impact loci showed association with all seven diseases. MVMR supported a causal role of ApoB in ischaemic heart disease after accounting for lipid species. Phosphatidylethanolamine-increasing LIPC variants seemed to lower age-related macular degeneration risk independently of HDL-cholesterol. MVMR suggested a protective effect of four lipid species containing arachidonic acid on cholelithiasis risk independently of Total Cholesterol. INTERPRETATION:Our study demonstrates how genetic predictors of lipid species can be utilised to gain insights into disease risk. We report potential links between lipid species and age-related macular degeneration and cholelithiasis risk, which can be explored for their utility in disease risk prediction and therapy. FUNDING:The funders had no role in the study design, data analyses, interpretation, or writing of this article.

Infections can lead to persistent symptoms and diseases such as shingles after varicella zoster or rheumatic fever after streptococcal infections. Similarly, severe acute respiratory syndrome coronavirus 2 (SARS‑CoV‑2) infection can result in long coronavirus disease (COVID), typically manifesting as fatigue, pulmonary symptoms and cognitive dysfunction. The biological mechanisms behind long COVID remain unclear. We performed a genome-wide association study for long COVID including up to 6,450 long COVID cases and 1,093,995 population controls from 24 studies across 16 countries. We discovered an association of FOXP4 with long COVID, independent of its previously identified association with severe COVID-19. The signal was replicated in 9,500 long COVID cases and 798,835 population controls. Given the transcription factor FOXP4's role in lung physiology and pathology, our findings highlight the importance of lung function in the pathophysiology of long COVID.

Background:Episodic memory (EM) refers to the ability to encode and recall events-a vital cognitive function for healthy cognitive aging and an endophenotype for dementia. Methods:Using genome- and metabolome-wide least absolute shrinkage and selection operator (LASSO) analysis, we developed polygenic (LASSO-PRS) and metabolic risk scores (MRS) in ∼68.5-year-old individuals (n = 897). We also applied the Bayesian regression method PRS-CS to an external genome-wide meta-analysis (GWAMA, N = 29,785, age > 18 years) to derive another PRS (GWAMA-PRS). We assessed incremental variances (R 2) in EM explained by the PRSs and MRS separately and in combination beyond the Cardiovascular Risk Factors, Aging, and Incidence of Dementia (CAIDE) score in 104 independent ∼68.5-year-old individuals. Finally, we validated the PRSs in 2 independent pediatric cohorts (N = 309, age = ∼11.9 years; N = 443, age = ∼8.6 years). Results:In the independent sample of ∼68.5-year-old individuals, compared with CAIDE score alone, accounting additionally for either MRS, LASSO-PRS, or GWAMA-PRS increased R 2 by 1.6, 5.6, and 4.5 percentage points (pp), respectively, while accounting additionally for MRS + LASSO-PRS or MRS + GWAMA-PRS increased R 2 by 7.8 and 6.4 pp, respectively. Both LASSO-PRS (all false discovery rate [FDR]-adjusted p values = .01-.03) and GWAMA-PRS (all FDR-adjusted p values = .03) were significantly associated with EM in all models, while the CAIDE score and MRS were not (all FDR-adjusted p values > .05). PRSs were not associated with EM in the pediatric cohorts (all FDR-adjusted p values > .05). Conclusions:Genomics added predictive value to EM beyond epidemiological risk factors in adults, but the same was not observed with metabolomics. Adult-derived PRSs did not predict EM in children.

Assessing how much variability in blood plasma proteins is due to genetic or environmental factors is essential for advancing personalized medicine. While large-scale studies have established SNP-based heritability (SNP-h2) estimates for plasma proteins, less is known about the proportion of total genetic effects on protein variability. We applied quantitative genetic twin models to estimate the heritability of 2321 plasma proteins and to assess the proportion of heritability accounted for by SNP-h2 estimates. Olink proteomics data were generated for 401 twins aged 56-70, including 196 complete same-sex twin pairs. On average, 40% of protein variability was attributable to genetic effects. Twin-based heritability estimates were highly correlated with published SNP-h2 estimates from the UK Biobank (Spearman coefficient: ρ = 0.80). However, on average, only half of the total heritability was covered by SNP-h2, and the other half, representing one-fifth of the total protein phenotypic variability, remains missing.

Recent statistical approaches have shown that the set of all available genetic variants explains considerably more phenotypic variance of complex traits and diseases than the individual variants that are robustly associated with these phenotypes. However, rapidly increasing sample sizes constantly improve detection and prioritization of individual variants driving the associations between genomic regions and phenotypes. Therefore, it is useful to routinely estimate how much phenotypic variance the detected variants explain for each region by taking into account the correlation structure of variants and the uncertainty in their causal status. Here we extend the FINEMAP software to estimate the effect sizes and regional heritability under the probabilistic model that assumes a handful of causal variants per region. Using the UK Biobank (UKB) data to simulate genomic regions, we demonstrate that FINEMAP provides higher precision and enables more detailed decomposition of regional heritability into individual variants than the variance component model implemented in BOLT or the fixed-effect model implemented in HESS, particularly when there are only a few causal variants in the fine-mapped region. Using data from 2,940 plasma proteins from the UKB study, we observed that on average FINEMAP identified 2.5 causal variants at an association signal and captured 36% more regional heritability than the variant with the lowest P-value. We estimate that in genomic regions with notable contribution to the total heritability, FINEMAP captures on average 13% and 40% more heritability than BOLT and HESS respectively. Our analysis shows how FINEMAP, BOLT and HESS relate to each other in cases where inference of a variant-level picture of the regional genetic architecture is possible.

The high prevalence of autoimmune hypothyroidism (AIHT) - more than 5% in human populations - provides a unique opportunity to unlock the most complete picture to date of genetic loci that underlie systemic and organ-specific autoimmunity. Using a meta-analysis of 81,718 AIHT cases in FinnGen and the UK Biobank, we dissect associations along axes of thyroid dysfunction and autoimmunity. This largest-to-date scan of hypothyroidism identifies 418 independent associations (p < 5×10- 8), more than half of which have not previously been documented in thyroid disease. In 48 of these, a protein-coding variant is the lead SNP or is highly correlated (r2 > 0.95) with the lead SNP at the locus, including low-frequency coding variants at LAG3, ZAP70, TG, TNFSF11, IRF3, S1PR4, HABP2, ZNF429 as well as established variants at ADCY7, IFIH1 and TYK2. The variants at LAG3 (P67T), ZAP70 (T155M), and TG (Q655X) are highly enriched in Finland and functional experiments in T-cells demonstrate that the ZAP70:T155M allele reduces T-cell activation. By employing a large-scale scan of non-thyroid autoimmunity and a published meta-analysis of TSH levels, we use a Bayesian classifier to dissect the associated loci into distinct groupings and from this estimate, a significant proportion are involved in systemic (i.e., general to multiple autoimmune conditions) autoimmunity (34%) and another subset in thyroid-specific dysfunction (17%). By comparing these association results further to other common disease endpoints, we identify a noteworthy overlap with skin cancer, with 10% of AIHT loci showing a consistent but opposite pattern of association where alleles that increase the risk of hypothyroidism have protective effects for skin cancer. The association results, including genes encoding checkpoint inhibitors and other genes affecting protein levels of PD1, bolster the causal role of natural variation in autoimmunity influencing cancer outcomes.

Epidemiological studies have robustly linked lower birth weight to later-life disease risks. These observations may reflect the adverse impact of intrauterine growth restriction on a child’s health. However, causal evidence supporting such a mechanism in humans is largely lacking. Using Mendelian Randomization and 36,211 genotyped mother-child pairs from the FinnGen study, we assessed the relationship between intrauterine growth and five common health outcomes (coronary heart disease (CHD), hypertension, statin use, type 2 diabetes and cancer). We proxied intrauterine growth with polygenic scores for maternal effects on birth weight and took into account the transmission of genetic variants between a mother and a child in the analyses. We find limited evidence for contribution of normal variation in maternally influenced intrauterine growth on later-life disease. Instead, we find support for genetic pleiotropy in the fetal genome linking birth weight to CHD and hypertension. Our study illustrates the opportunities that data from genotyped parent-child pairs from a population-based biobank provides for addressing causality of maternal influences.

Abstract Background Many histopathologic prognostic markers, identified by routine hematoxylin and eosin (HE) staining, have been proposed for predicting the survival of patients with oral squamous cell carcinoma (OSCC). Subsequently, several meta-analyses have been conducted on these prognostic markers. We sought to analyze the accumulated evidence from these meta-analyses. Methods An electronic database search of PubMed, Scopus, Ovid Medline, Web of Science, and Cochrane Library was conducted to retrieve all meta-analysis articles published on histopathologic prognostic markers of OSCC. The risk of bias of the included studies was analyzed using the Risk of Bias in Systematic Reviews (ROBIS) tool. The synthesis of the results was conducted following the guidelines of Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA). Results There were 16 meta-analysis articles published on the histological prognostic markers of OSSC. The accumulated evidence from these meta-analyses highlighted the powerful prognostic value of depth of invasion, tumor thickness, perineural invasion, lymphovascular invasion, worst pattern of invasion, tumor budding, and tumor-stroma ratio. The highest odds ratio (OR) of a relationship between a histopathologic prognostic marker and outcome was for the depth of invasion (OR 10.16, 95% CI 5.05–20.46) and tumor thickness (OR 7.32, 95% CI 5.3–10.1) in predicting lymph node metastasis. Conclusion The published meta-analyses present robust evidence on the significance of emerging histopathologic markers, namely, worst pattern of invasion, tumor budding, and tumor-stroma ratio. It is time to consider such markers in daily pathology reporting and risk stratification of OSCC.

Gestational diabetes mellitus (GDM) is a common metabolic disorder affecting more than 16 million pregnancies annually worldwide 1 , 2 . GDM is related to an increased lifetime risk of type 2 diabetes (T2D) 1 – 3 , with over a third of women developing T2D within 15 years of their GDM diagnosis. The diseases are hypothesized to share a genetic predisposition 1 – 7 , but few studies have sought to uncover the genetic underpinnings of GDM. Most studies have evaluated the impact of T2D loci only 8 – 10 , and the three prior genome-wide association studies of GDM 11 – 13 have identified only five loci, limiting the power to assess to what extent variants or biological pathways are specific to GDM. We conducted the largest genome-wide association study of GDM to date in 12,332 cases and 131,109 parous female controls in the FinnGen study and identified 13 GDM-associated loci, including nine new loci. Genetic features distinct from T2D were identified both at the locus and genomic scale. Our results suggest that the genetics of GDM risk falls into the following two distinct categories: one part conventional T2D polygenic risk and one part predominantly influencing mechanisms disrupted in pregnancy. Loci with GDM-predominant effects map to genes related to islet cells, central glucose homeostasis, steroidogenesis and placental expression.

Background Machine learning (ML) classifiers are increasingly used for predicting cardiovascular disease (CVD) and related risk factors using omics data, although these outcomes often exhibit categorical nature and class imbalances. However, little is known about which ML classifier, omics data, or upstream dimension reduction strategy has the strongest influence on prediction quality in such settings. Our study aimed to illustrate and compare different machine learning strategies to predict CVD risk factors under different scenarios.Methods We compared the use of six ML classifiers in predicting CVD risk factors using blood-derived metabolomics, epigenetics and transcriptomics data. Upstream omic dimension reduction was performed using either unsupervised or semi-supervised autoencoders, whose downstream ML classifier performance we compared. CVD risk factors included systolic and diastolic blood pressure measurements and ultrasound-based biomarkers of left ventricular diastolic dysfunction (LVDD; E/e' ratio, E/A ratio, LAVI) collected from 1,249 Finnish participants, of which 80% were used for model fitting. We predicted individuals with low, high or average levels of CVD risk factors, the latter class being the most common. We constructed multi-omic predictions using a meta-learner that weighted single-omic predictions. Model performance comparisons were based on the F1 score. Finally, we investigated whether learned omic representations from pre-trained semi-supervised autoencoders could improve outcome prediction in an external cohort using transfer learning.Results Depending on the ML classifier or omic used, the quality of single-omic predictions varied. Multi-omics predictions outperformed single-omics predictions in most cases, particularly in the prediction of individuals with high or low CVD risk factor levels. Semi-supervised autoencoders improved downstream predictions compared to the use of unsupervised autoencoders. In addition, median gains in Area Under the Curve by transfer learning compared to modelling from scratch ranged from 0.09 to 0.14 and 0.07 to 0.11 units for transcriptomic and metabolomic data, respectively.Conclusions By illustrating the use of different machine learning strategies in different scenarios, our study provides a platform for researchers to evaluate how the choice of omics, ML classifiers, and dimension reduction can influence the quality of CVD risk factor predictions.

Assessing how much of the variability in blood plasma proteins is due to genetic or environmental factors is essential for advancing personalized medicine. While large-scale studies have established SNP-based heritability (SNP-h2) estimates for plasma proteins, less is known about the proportion of total genetic effects on protein variability. We applied quantitative genetic twin models to estimate the heritability of 2,321 plasma proteins and to assess the proportion of heritability accounted for by SNP-h2 estimates. Olink proteomics data were generated for 401 twins aged 56-70, including 196 complete same-sex twin pairs. On average, 40% of protein variability was attributable to genetic effects. Twin-based heritability estimates were highly correlated with published SNP-h2 estimates from the UK Biobank (Spearman coefficient: r=0.80). However, on average, only half of the total heritability was covered by SNP-h2, and the other half, representing one-fifth of total protein phenotypic variability, remains missing.### Competing Interest StatementThe authors have declared no competing interest.### Funding StatementData collection of the EH-Epi sample has been supported by the Academy of Finland (Grants 265240, 263278, 308248, 312073, 336832 to JK and 297908 to MO) and the Sigrid Juselius Foundation (to MO). We want to acknowledge the participants and investigators of FinnGen study. The FinnGen project is funded by two grants from Business Finland (HUS 4685/31/2016 and UH 4386/31/2016) and the following industry partners: AbbVie Inc., AstraZeneca UK Ltd, Biogen MA Inc., Bristol Myers Squibb (and Celgene Corporation & Celgene International II Sarl), Genentech Inc., Merck Sharp & Dohme LCC, Pfizer Inc., GlaxoSmithKline Intellectual Property Development Ltd., Sanofi US Services Inc., Maze Therapeutics Inc., Janssen Biotech Inc, Novartis AG, and Boehringer Ingelheim International GmbH. Following biobanks are acknowledged for delivering biobank samples to FinnGen: Auria Biobank (www.auria.fi/biopankki), THL Biobank (www.thl.fi/biobank), Helsinki Biobank (www.helsinginbiopankki.fi), Biobank Borealis of Northern Finland (https://www.ppshp.fi/Tutkimus-ja-opetus/Biopankki/Pages/Biobank-Borealis-briefly-in-English.aspx), Finnish Clinical Biobank Tampere (www.tays.fi/en-US/Research\_and\_development/Finnish\_Clinical\_Biobank_Tampere), Biobank of Eastern Finland (www.ita-suomenbiopankki.fi/en), Central Finland Biobank (www.ksshp.fi/fi-FI/Potilaalle/Biopankki), Finnish Red Cross Blood Service Biobank (www.veripalvelu.fi/verenluovutus/biopankkitoiminta), Terveystalo Biobank (www.terveystalo.com/fi/Yritystietoa/Terveystalo-Biopankki/Biopankki/) and Arctic Biobank (https://www.oulu.fi/en/university/faculties-and-units/faculty-medicine/northern-finland-birth-cohorts-and-arctic-biobank). All Finnish Biobanks are members of BBMRI.fi infrastructure (www.bbmri.fi). Finnish Biobank Cooperative -FINBB (https://finbb.fi/) is the coordinator of BBMRI-ERIC operations in Finland. The Finnish biobank data can be accessed through the Fingenious services (https://site.fingenious.fi/en/) managed by FINBB.### Author DeclarationsI confirm all relevant ethical guidelines have been followed, and any necessary IRB and/or ethics committee approvals have been obtained.YesThe details of the IRB/oversight body that provided approval or exemption for the research described are given below:The study protocol was approved by the Institutional Ethics Board of the Hospital District of Helsinki and Uusimaa, Finland (ID 154/13/03/00/11) and the Institutional Review Board of Augusta UniversityI confirm that all necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived, and that any patient/participant/sample identifiers included were not known to anyone (e.g., hospital staff, patients or participants themselves) outside the research group so cannot be used to identify individuals.YesI understand that all clinical trials and any other prospective interventional studies must be registered with an ICMJE-approved registry, such as ClinicalTrials.gov. I confirm that any such study reported in the manuscript has been registered and the trial registration ID is provided (note: if posting a prospective study registered retrospectively, please provide a statement in the trial ID field explaining why the study was not registered in advance).YesI have followed all appropriate research reporting guidelines, such as any relevant EQUATOR Network research reporting checklist(s) and other pertinent material, if applicable.YesThe Finnish Twin Cohort data used in the analysis is deposited in the Biobank of the Finnish Institute for Health and Welfare (https://thl.fi/en/web/thl-biobank/forresearchers). It is available to researchers after written application and following the relevant Finnish legislation.

PURPOSE Family history (FH) and pathogenic variants (PVs) are used for guiding risk surveillance in selected high-risk women but little is known about their impact for breast cancer screening on population level. In addition, polygenic risk scores (PRSs) have been shown to efficiently stratify breast cancer risk through combining information about common genetic factors into one measure. METHODS In longitudinal real-life data, we evaluate PRS, FH, and PVs for stratified screening. Using FinnGen (N = 117,252), linked to the Mass Screening Registry for breast cancer (1992-2019; nationwide organized biennial screening for age 50-69 years), we assessed the screening performance of a breast cancer PRS and compared its performance with FH of breast cancer and PVs in moderate- ( CHEK2 )- to high-risk ( PALB2 ) susceptibility genes. RESULTS Effect sizes for FH, PVs, and high PRS (>90th percentile) were comparable in screening-aged women, with similar implications for shifting age at screening onset. A high PRS identified women more likely to be diagnosed with breast cancer after a positive screening finding (positive predictive value [PPV], 39.5% [95% CI, 37.6 to 41.5]). Combinations of risk factors increased the PPVs up to 45% to 50%. A high PRS conferred an elevated risk of interval breast cancer (hazard ratio [HR], 2.78 [95% CI, 2.00 to 3.86] at age 50 years; HR, 2.48 [95% CI, 1.67 to 3.70] at age 60 years), and women with a low PRS (<10th percentile) had a low risk for both interval- and screen-detected breast cancers. CONCLUSION Using real-life screening data, this study demonstrates the effectiveness of a breast cancer PRS for risk stratification, alone and combined with FH and PVs. Further research is required to evaluate their impact in a prospective risk-stratified screening program, including cost-effectiveness.