Prioritizing context-specific genetic risk mechanisms in 11 solid cancers

Prioritizing context-specific genetic risk mechanisms in 11 solid cancers | medRxiv /* */ /* */ <!-- <!-- /*! * yepnope1.5.4 * (c) WTFPL, GPLv2 */ (function(a,b,c){function d(a){return"[object Function]"==o.call(a)}function e(a){return"string"==typeof a}function f(){}function g(a){return!a||"loaded"==a||"complete"==a||"uninitialized"==a}function h(){var a=p.shift();q=1,a?a.t?m(function(){("c"==a.t?B.injectCss:B.injectJs)(a.s,0,a.a,a.x,a.e,1)},0):(a(),h()):q=0}function i(a,c,d,e,f,i,j){function k(b){if(!o&&g(l.readyState)&&(u.r=o=1,!q&&h(),l.onload=l.onreadystatechange=null,b)){"img"!=a&&m(function(){t.removeChild(l)},50);for(var d in y[c])y[c].hasOwnProperty(d)&&y[c][d].onload()}}var j=j||B.errorTimeout,l=b.createElement(a),o=0,r=0,u={t:d,s:c,e:f,a:i,x:j};1===y[c]&&(r=1,y[c]=[]),"object"==a?l.data=c:(l.src=c,l.type=a),l.width=l.height="0",l.onerror=l.onload=l.onreadystatechange=function(){k.call(this,r)},p.splice(e,0,u),"img"!=a&&(r||2===y[c]?(t.insertBefore(l,s?null:n),m(k,j)):y[c].push(l))}function j(a,b,c,d,f){return q=0,b=b||"j",e(a)?i("c"==b?v:u,a,b,this.i++,c,d,f):(p.splice(this.i++,0,a),1==p.length&&h()),this}function k(){var a=B;return a.loader={load:j,i:0},a}var l=b.documentElement,m=a.setTimeout,n=b.getElementsByTagName("script")[0],o={}.toString,p=[],q=0,r="MozAppearance"in l.style,s=r&&!!b.createRange().compareNode,t=s?l:n.parentNode,l=a.opera&&"[object Opera]"==o.call(a.opera),l=!!b.attachEvent&&!l,u=r?"object":l?"script":"img",v=l?"script":u,w=Array.isArray||function(a){return"[object Array]"==o.call(a)},x=[],y={},z={timeout:function(a,b){return b.length&&(a.timeout=b[0]),a}},A,B;B=function(a){function b(a){var a=a.split("!"),b=x.length,c=a.pop(),d=a.length,c={url:c,origUrl:c,prefixes:a},e,f,g;for(f=0;f<d;f++)g=a[f].split("="),(e=z[g.shift()])&&(c=e(c,g));for(f=0;f<b;f++)c=x[f](c);return c}function g(a,e,f,g,h){var i=b(a),j=i.autoCallback;i.url.split(".").pop().split("?").shift(),i.bypass||(e&&(e=d(e)?e:e[a]||e[g]||e[a.split("/").pop().split("?")[0]]),i.instead?i.instead(a,e,f,g,h):(y[i.url]?i.noexec=!0:y[i.url]=1,f.load(i.url,i.forceCSS||!i.forceJS&&"css"==i.url.split(".").pop().split("?").shift()?"c":c,i.noexec,i.attrs,i.timeout),(d(e)||d(j))&&f.load(function(){k(),e&&e(i.origUrl,h,g),j&&j(i.origUrl,h,g),y[i.url]=2})))}function h(a,b){function c(a,c){if(a){if(e(a))c||(j=function(){var a=[].slice.call(arguments);k.apply(this,a),l()}),g(a,j,b,0,h);else if(Object(a)===a)for(n in m=function(){var b=0,c;for(c in a)a.hasOwnProperty(c)&&b++;return b}(),a)a.hasOwnProperty(n)&&(!c&&!--m&&(d(j)?j=function(){var a=[].slice.call(arguments);k.apply(this,a),l()}:j[n]=function(a){return function(){var b=[].slice.call(arguments);a&&a.apply(this,b),l()}}(k[n])),g(a[n],j,b,n,h))}else!c&&l()}var h=!!a.test,i=a.load||a.both,j=a.callback||f,k=j,l=a.complete||f,m,n;c(h?a.yep:a.nope,!!i),i&&c(i)}var i,j,l=this.yepnope.loader;if(e(a))g(a,0,l,0);else if(w(a))for(i=0;i (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0];var j=d.createElement(s);var dl=l!='dataLayer'?'&l='+l:'';j.src='//www.googletagmanager.com/gtm.js?id='+i+dl;j.type='text/javascript';j.async=true;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-P4HH5NV'); Skip to main content Home About Submit ALERTS / RSS Search for this keyword Advanced Search Prioritizing context-specific genetic risk mechanisms in 11 solid cancers Xueyao Wu , Artem Kim , Charles E Breeze , Tracy A O’Mara , Dhanya Ramachandran , Thilo Dörk , Stella Koutros , Nathaniel Rothman , View ORCID Profile Ludmila Prokunina-Olsson , Nicholas Mancuso , Sara Lindström , Peter Kraft doi: https://doi.org/10.1101/2025.10.30.25339145 Xueyao Wu 1 Trans-Divisional Research Program, Division of Cancer Epidemiology and Genetics, National Cancer Institute , Rockville, MD, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site Artem Kim 2 Department of Population and Public Health Sciences, Keck School of Medicine, University of Southern California , Los Angeles, CA, USA 3 Center for Genetic Epidemiology, Keck School of Medicine, University of Southern California , Los Angeles, CA, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site Charles E Breeze 4 Occupational and Environmental Epidemiology Branch, Division of Cancer Epidemiology and Genetics, National Cancer Institute , Rockville, MD, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site Tracy A O’Mara 5 Cancer Research Program , QIMR Berghofer, Brisbane, Queensland, Australia 6 Faculty of Health, Medicine and Behavioural Sciences, The University of Queensland , Brisbane, Queensland, Australia Find this author on Google Scholar Find this author on PubMed Search for this author on this site Dhanya Ramachandran 7 Gynaecology Research Unit, Hannover Medical School , Hannover, Germany Find this author on Google Scholar Find this author on PubMed Search for this author on this site Thilo Dörk 7 Gynaecology Research Unit, Hannover Medical School , Hannover, Germany Find this author on Google Scholar Find this author on PubMed Search for this author on this site Stella Koutros 4 Occupational and Environmental Epidemiology Branch, Division of Cancer Epidemiology and Genetics, National Cancer Institute , Rockville, MD, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site Nathaniel Rothman 4 Occupational and Environmental Epidemiology Branch, Division of Cancer Epidemiology and Genetics, National Cancer Institute , Rockville, MD, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site Ludmila Prokunina-Olsson 8 Laboratory of Translational Genomics, Division of Cancer Epidemiology and Genetics, National Cancer Institute , Rockville, MD, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Ludmila Prokunina-Olsson Nicholas Mancuso 2 Department of Population and Public Health Sciences, Keck School of Medicine, University of Southern California , Los Angeles, CA, USA 3 Center for Genetic Epidemiology, Keck School of Medicine, University of Southern California , Los Angeles, CA, USA 9 Department of Quantitative and Computational Biology, University of Southern California , Los Angeles, CA, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site Sara Lindström 10 Department of Epidemiology, School of Public Health, University of Washington , Seattle, WA, USA 11 Public Health Sciences Division, Fred Hutchinson Cancer Center , Seattle, WA, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site Peter Kraft 1 Trans-Divisional Research Program, Division of Cancer Epidemiology and Genetics, National Cancer Institute , Rockville, MD, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site For correspondence: phillip.kraft{at}nih.gov Abstract Full Text Info/History Metrics Supplementary material Preview PDF Abstract Background While genome-wide association studies (GWAS) have identified hundreds of cancer-associated genetic variants, the specific biological contexts where these variants exert their effects remain largely unknown. We aimed to prioritize context-specific genetic risk mechanisms for 11 solid cancers at both genome-wide and single-variant resolutions. Methods We integrated cancer GWAS summary statistics from European ancestry samples (avg. n cases=47,856) with ∼1,500 context-specific annotations representing candidate cis -regulatory elements. For genome-wide analysis, we applied CT-FM, a method that leverages heritability enrichment estimates and an annotation correlation matrix to select likely disease-relevant biological contexts. After identifying putative causal SNPs (PIP≥0.5) via functionally informed fine-mapping, we used CT-FM-SNP to identify relevant contexts for individual variants. A combined SNP-to-gene framework was applied to construct putative {regulatory SNP-context-gene-cancer} quadruplets. Results Stratified LD score regression analysis identified 52 annotations with significant heritability enrichment (Bonferroni-corrected P ≤0.05). CT-FM prioritized four high-confidence (PIP≥0.5) biological contexts: mammary luminal epithelial cells for breast cancer, a prostate cancer epithelial cell line (VCaP) for prostate cancer, and bulk tumor tissue contexts for colorectal and renal cancers. Variant-level analysis of hundreds of putatively causal SNPs corroborated these findings and identified additional high-confidence contexts for other malignancies, including estrogen receptor-negative breast cancer and bladder cancer. A total of 489 putative regulatory quadruplets were constructed, proposing specific molecular mechanisms underlying the observed GWAS signals. Conclusion These findings advance our understanding of genetic susceptibility to different cancers. Future work in larger, more diverse GWAS, coupled with more comprehensive annotation atlases, is essential to expand upon and validate our results. Introduction Cancer remains a leading cause of morbidity and mortality worldwide, with a global burden that is expected to rise significantly in the coming decades. This highlights the urgent need for improved prevention, early detection, and treatment strategies. 1 , 2 While genome-wide association studies (GWAS) have identified hundreds of genetic variants associated with cancer susceptibility, the vast majority of these variants are located in non-coding regions of the genome. 3 Translating these associations into biological function and clinical relevance presents a formidable challenge, central to which is pinpointing the causal variants and understanding their context-specific regulatory mechanisms. 4 Cancer is driven by complex interactions among diverse cell populations. 5 Genetic risk variants likely exert their effects by perturbing gene regulatory networks within specific contexts, 6 , 7 ranging from individual cell types to complex tissue environments comprising multiple cell populations, such as the tumor microenvironment (TME). 8 For instance, breast cancer risk variant rs2981578 near the FGFR2 gene selectively alters the binding of FOXA1, a pioneer factor critical for estrogen receptor (ER) function expressed in only a subset of luminal epithelial cells within the human mammary gland. 9 Analysis of thousands of lung cancer variants revealed that a majority of the risk variants found in regulatory regions were specific to a single cell-type category, with epithelial and immune cells emerging as the most significant contributors to lung cancer susceptibility. 10 Understanding the biological contexts where cancer-related genetic associations individually and collectively alter gene regulation is therefore critical for unraveling the cellular origins of cancer, prioritizing experimental validation studies, and informing the development of targeted therapies. 11 Large-scale functional genomics initiatives, including the ENCyclopedia of DNA Elements (ENCODE4) and The Cancer Genome Atlas (TCGA) projects, 12 , 13 have generated comprehensive epigenomic profiles across diverse tissue and cellular environments, creating opportunities for integration with GWAS data to elucidate context-specific mechanisms in cancer genetics. Statistical methods such as Stratified Linkage Disequilibrium Score Regression (S-LDSC) 14 , 15 leverage these functional annotations for partitioning heritability and identifying disease-relevant biological contexts. However, when evaluating each annotation independently, S-LDSC cannot distinguish truly causal from ‘tagging’ associations arising from context-shared regulatory patterns (e.g., different immune cell types or epithelial cells from different tissues). 16 , 17 Consequently, it frequently identifies multiple, highly correlated annotations as statistically significant, substantially obscuring the primary drivers of heritability signals. 18 On the other hand, performing S-LDSC while mutually adjusting for a large number of context-specific annotations is computationally intractable. This challenge underscores the need for computational frameworks that explicitly model shared regulatory architectures across diverse contexts to enable more precise identification of the biological processes driving cancer development. 19 , 20 To address these limitations, we implemented an integrative statistical fine-mapping framework to systematically prioritize the biological contexts underlying genetic risk for 11 solid cancers. Our approach centers on Cell-Type Fine-Mapping (CT-FM) and its single-variant extension, CT-FM-SNP, which adapt established SNP fine-mapping principles to the prioritization of cell-type and tissue contexts by jointly modeling heritability enrichments across annotations. 20 While these methods have demonstrated success in prioritizing cellular contexts and disentangling correlated signals for multiple continuous traits, their application to cancer and other complex diseases remains limited. To further enhance precision, we employed an advanced, functionally informed SNP fine-mapping strategy to generate refined sets of putative causal variants for each cancer type, which serve as high-confidence input for CT-FM-SNP analysis. 21 By linking these context-specific variants to their putative target genes, we constructed end-to-end hypotheses that specify the likely causal SNP, its functional context, the gene it regulates, and the associated cancer. Through the integration of these complementary approaches, we sought to dissect the genetic architecture of cancer susceptibility, identify putative regulatory mechanisms, and establish a foundation for future experimental validation and therapeutic target discovery. The overall study design is presented in Figure 1 . Download figure Open in new tab Figure 1. Overall study design. This schematic illustrates the integrative statistical fine-mapping framework employed to prioritize context-specific genetic risk mechanisms in 11 solid cancers. The workflow begins with the integration of large-scale GWAS summary statistics and an atlas of ∼1,500 context-specific cis -regulatory annotations. Cell-Type Fine-Mapping (CT-FM) is applied for genome-wide prioritization of biological contexts. Concurrently, functionally informed SNP fine-mapping identifies high-confidence causal variants. These variants then serve as input for CT-FM-SNP to pinpoint relevant contexts at a single-variant resolution. Finally, a combined SNP-to-gene (cS2G) framework links context-specific variants to their putative target genes, culminating in the construction of {regulatory SNP-context-gene-cancer} quadruplets. Materials and Methods Context-specific SNP-annotations Our dataset includes 1,473 SNP annotations, each defining a candidate cis -regulatory element in a specific biological context ( Table S1 ). This collection was assembled by first incorporating 927 cell-type-specific annotations previously curated for the original CT-FM framework from the fourth phase of ENCODE (ENCODE4), 22 Cis-element Atlas (CATlas), 23 and Activity-by-Contact (ABC) model. 24 We then expanded this library with 523 additional cell-type-specific annotations, including cell lines and cancer-related cell types from the original consortia to enhance the scope for cancer-specific analysis, as well as annotations from Epigenome Integration across Multiple Annotation Projects (Epimap) 25 to broaden overall cellular diversity. We also integrated 23 bulk tumor annotations from TCGA. 13 Although cell-type-specific data from primary tumors are not widely available, these bulk tumor profiles are essential as they capture the aggregate regulatory landscape of the in vivo tumor ecosystem. Our final library therefore combines the cellular specificity of primary cells and cell lines (n = 1,450) with the disease relevance of primary tumor tissues (n = 23). Detailed information on annotation data types, curation and quality control procedures is available in the Supplementary Methods . GWAS summary statistics We assembled GWAS summary statistics from European ancestry samples for 11 cancer types. The datasets encompassed breast cancer (total n=228,951; 122,977 cases), with subtype-specific data for estrogen receptor-positive (ER+; 69,501 cases) and estrogen receptor-negative (ER–; 21,468 cases) breast cancer; endometrial cancer (n= 244,644; 16,078 cases); ovarian cancer (n=63,347; 22,406 cases); prostate cancer (n=726,828; 122,188 cases); colorectal cancer (n=185,886; 78,743 cases); esophageal cancer (n=21,271; 4,112 cases); lung cancer (n=85,716; 29,266 cases); renal cancer (n=769,475; 25,890 cases); and bladder cancer (n=357,292; 13,790 cases). Most cancer GWAS datasets were imputed using 1000 Genomes Project (1KGP) reference panels. Study characteristics and participant information are summarized in Table 1 with further details available in the original cancer-specific publications. 26 – 34 View this table: View inline View popup Download powerpoint Table 1. Overview of cancer GWAS datasets. We applied the following quality control filters to each GWAS dataset. We removed variants on sex chromosomes, multi-allelic SNPs, and ambiguous variants; variants with low imputation quality (INFO score < 0.9), low minor allele frequency [MAF] 3), or invalid P -values ( P -value ≤ 0 or > 1); and variants sharing identical physical positions. We harmonized all datasets to the hg19 genome build. Finally, we aligned the datasets with the 1KGP v3 European reference panel, excluding variants absent from the panel, variants with MAF 0.2) between the GWAS and the reference panel population. CT-FM analysis We utilized CT-FM to prioritize biological contexts underlying cancer-specific GWAS signals. 20 CT-FM first applies S-LDSC, using LD scores from the 1KGP v3 European reference panel and adjusting for baseline genomic features, to estimate marginal effects of each context-specific annotation on SNP heritability and generate a genome-wide heritability enrichment Z-score for each cancer-context pair. 15 We used 1/(1/Ncase + 1/Ncontrol) as the effective sample size in the S-LDSC analysis (see rationale in the supplementary note of 35 ). Using enrichment estimates with positive z-scores and a correlation matrix of these annotations (i.e., correlations between the baseline-adjusted LD scores of context-specific annotations) 20 , CT-FM then employs the Sum of Single Effects (SuSiE) model to fine-map annotations, assuming a maximum of 10 independent causal contexts (L = 10, default setting of CT-FM). 36 , 37 This process generates (i) annotation-specific posterior inclusion probabilities (PIPs), reflecting the probability of an annotation being a primary driver of heritability enrichment, and (ii) credible sets, where each set represents an independent signal consisting of the smallest group of correlated annotations (with a minimum absolute pairwise correlation ≥ 0.192, the average correlation of the annotation matrix) that together have a 95% probability of including the true causal context. We identified annotations showing statistically significant enrichment in S-LDSC using both Bonferroni-corrected and FDR-corrected P -value thresholds of 0.05 within each cancer. Following the original CT-FM framework, we defined high-confidence biological contexts as those with PIPs ≥ 0.5. To increase our sensitivity to identify candidate biological contexts, we also used a more lenient threshold of 0.05 ≤ PIPs < 0.5. SNP-fine-mapping analysis We conducted functionally informed SNP fine-mapping using PolyFun integrated with SuSiE to identify high-confidence causal variants for each cancer. 21 PolyFun leverages genome-wide functional annotations including coding, conserved, regulatory, and LD-related genomic features to compute prior causal probabilities for variants, proportional to their predicted per-SNP heritability contributions. These annotation-derived priors were then incorporated into the SuSiE fine-mapping algorithm to improve causal variant identification. For our analysis, we utilized a set of pre-computed PolyFun prior probabilities derived from a meta-analysis of 15 genetically uncorrelated UK Biobank traits, covering approximately 19 million imputed SNPs with MAF > 0.1%. Using these functional priors, we applied SuSiE fine-mapping to overlapping 3MB regions across the genome, restricting our analysis to those containing at least one genome-wide significant (P < 5ξ10 −8 ) SNP and assuming 5 causal SNPs in each region (L = 5, default setting of PolyFun). To resolve duplicate analyses for SNPs present in multiple regions, we retained only the result from the region where the SNP was most centrally located. The 1KGP European reference panel was used for LD information. Effective sample size was calculated as Neff = 4/(1/Ncase + 1/Ncontrol), following the approach adopted by the original publication. 21 High-confidence causal variants were defined with a PIP threshold of ≥ 0.50. The resulting set of variants was then used as input for CT-FM-SNP. CT-FM-SNP analysis To link specific candidate variants to their most relevant biological contexts, we further implemented CT-FM-SNP for each cancer. 20 CT-FM-SNP differs from CT-FM by restricting the SuSiE fine-mapping analysis to subsets of context-specific annotations that overlap with a given variant. We used the cancer-specific high-confidence causal variants identified through PolyFun+SuSiE fine-mapping as inputs. CT-FM-SNP first identified the subset of annotations that overlapped with the variant and then extracted the genome-wide heritability enrichment estimates for these annotations, as previously calculated by CT-FM. For variants overlapping at least two annotations with positive enrichment z-scores, SuSiE fine-mapping was applied using these variant-restricted annotations and the corresponding annotation correlation matrix, outputting variant-specific PIPs and credible sets for each annotation-cancer pair. As previously, high-confidence biological contexts were defined as those with PIPs ≥ 0.5. We did not focus on candidate contexts at the single-variant level given that the model was restricted to annotations with direct SNP overlap, a scenario where a lenient threshold might elevate the risk of false-positive findings. Identification of {regulatory SNP-context-gene-cancer} quadruplets To generate specific testable hypotheses, we integrated our SNP-level context results with a comprehensive SNP-to-gene (S2G) mapping framework to identify {regulatory SNP-context-gene-cancer} quadruplets. Our approach employed the combined S2G (cS2G) framework which integrates multiple S2G linking strategies to identify putative target genes for regulatory SNPs. 38 The cS2G framework incorporates 10 functionally informed S2G strategies, such as exon-based, promoter-based, fine-mapped cis -eQTLs, enhancer-gene linking, and chromatin interaction data, to assign linking scores to SNP-gene pairs. For our analysis, we utilized pre-computed cS2G linking scores derived from the 1KGP European reference panel. To form putative quadruplets, we systematically linked the SNP-context pairs that passed our high-confidence threshold (CT-FM-SNP PIP ≥ 0.5) with high-confidence (cS2G score ≥ 0.5) SNP-gene pairs from the cS2G framework. Sensitivity analyses excluding bulk tumor annotations To assess the robustness of our findings in the absence of bulk tumor annotations, we performed sensitivity analyses by repeating all analyses (CT-FM, CT-FM-SNP, and quadruplet construction) after excluding the 23 TCGA bulk tumor annotations. Results Identification of context-specific heritability enrichment In total, S-LDSC identified 52 biological contexts showing significant heritability enrichment (Bonferroni-corrected P ≤ 0.05) across five cancer types: overall breast cancer, ER+ breast cancer, prostate cancer, colorectal cancer, and renal cancer. At a less stringent significance threshold (q-value ≤ 0.05), S-LDSC identified 19, 13, 82, 22, and 5 contexts for overall breast, ER+ breast, prostate, colorectal, and renal cancers, respectively, totaling 141 contexts ( Figure 2 ). Download figure Open in new tab Figure 2. Statistical prioritization of biological contexts across six solid cancer types. Bar chart showing the number of annotations identified for six cancer types (breast, ER+ breast, prostate, colorectal, renal, and bladder) that demonstrated significant genome-wide enrichment signals. Annotations are categorized by increasing levels of statistical stringency: from annotations with stratified linkage disequilibrium score regression (S-LDSC) false discovery rate (FDR) q-value ≤ 0.05 to those meeting Bonferroni-corrected significance ( P ≤ 0.05) in S-LDSC analysis; and from candidate biological contexts with Cell-Type Fine-Mapping (CT-FM) posterior inclusion probability (PIP) ≥ 0.05 to high-confidence contexts with CT-FM PIP ≥ 0.5. Fine-mapping biological contexts at genome-wide resolution CT-FM substantially refined the initial S-LDSC results ( Table 2 , Figure 2 ), identifying four contexts achieved high confidence (PIPs ≥ 0.5): mammary luminal epithelial cells for both overall and ER+ breast cancer (PIP = 0.62-0.75), VCaP prostate adenocarcinoma epithelial cell line for prostate cancer (PIP = 1.00), colon adenocarcinoma (COAD) bulk tissue for colorectal cancer (PIP = 1.00), and kidney renal clear cell carcinoma (KIRC) bulk tissue for renal cancer (PIP = 0.73). View this table: View inline View popup Download powerpoint Table 2. High-confidence (PIP ≥ 0.5) and candidate (PIP ≥ 0.05) biological contexts identified by genome-wide context fine-mapping. Beyond these top signals, CT-FM identified 11 candidate (0.05 ≤ PIPs < 0.5) contexts across six cancers ( Table 2 ). These included mammary basal epithelial cells in overall and ER+ breast cancer, and T47D mammary luminal adenocarcinoma epithelial cell line in ER+ breast cancer. Prostate cancer analysis revealed prostate adenocarcinoma (PRAD) bulk tissue, while colorectal cancer analyses identified stromal contributions from connective tissue mesenchymal stem cells and various fibroblast populations. Renal cancer analysis implicated peripheral blood naïve B cells and renal proximal tubule epithelial cells. For bladder cancer, fetal ureteric bud epithelial cells and RWPE-2 prostate epithelial non-malignant cell line emerged as candidate contexts, despite bladder cancer lacking significant S-LDSC associations. A single 95% credible set was identified for each of these cancers except for bladder cancer, for which no credible set was formed given no combination of annotations reached the required 95% posterior probability threshold. Detailed S-LDSC and CT-FM results for all annotations satisfying the S-LDSC q-value ≤ 0.05 or the CT-FM PIP ≥ 0.05 criterion are provided in Table S2 . Fine-mapping biological contexts at single-variant resolution A total of 837 high-confidence (PIPs ≥ 0.5) causal SNPs were identified by functionally informed SNP fine-mapping across all cancer types except esophageal cancer. Among these, 487 SNPs overlapping at least two annotations with positive enrichment z-scores were further tested (counts by cancer in Table 3 ). View this table: View inline View popup Download powerpoint Table 3. Summary of high-confidence (PIP ≥ 0.5) biological contexts identified by variant-level context fine-mapping. We performed biological context fine-mapping by focusing on each of these 487 SNPs and restricting analyses to their overlapping annotations through CT-FM-SNP. This variant-level approach yielded altogether 639 high-confidence (PIPs ≥ 0.5) SNP-context-cancer triplets across nine of the 11 cancer types analyzed, excluding ovarian and esophageal cancers (full list in Table S3 ). Prostate cancer yielded the greatest number of high-confidence associations (322 triplets involving 81 unique biological contexts), followed by overall breast cancer (119 triplets, 53 contexts), ER+ breast cancer (92 triplets, 46 contexts), and renal cancer (59 triplets, 43 contexts). The variant-level signals were largely consistent with the genome-wide findings while capturing additional relevant high-confidence contexts ( Table 3 and Table S3 ). For both overall and ER+ breast cancer, mammary luminal epithelial cells represented the most frequently detected context, achieving high confidence in the analysis of 22 and 20 SNPs, respectively. ER– breast cancer predominantly implicated breast invasive carcinoma (BRCA) bulk tissue, although this association was observed for only 2 SNPs. Prostate cancer identified PRAD as the most frequent context across 86 SNPs, followed by VCaP across 85 SNPs. Colorectal cancer implicated COAD for 2 SNPs, while renal cancer most frequently identified KIRC across 17 SNPs. Bladder cancer demonstrated fetal ureteric bud epithelial cells as the most frequent context, detected for 2 SNPs. Endometrial and lung cancer yielded three and 13 high-confidence contexts, respectively—including COAD for endometrial and the MM.1S multiple myeloma cell line for lung—with each context detected for only one SNP. Among 53 SNPs overlapping only one annotation with positive z-score, none were linked to high-confidence contexts detected in the genome-wide CT-FM analysis, though three were linked to candidate contexts (rs61752561, rs61735792, and rs7591218; all for prostate cancer and overlapping PRAD). {Regulatory SNP-context-gene-cancer} quadruplets By integrating the SNP-context-cancer triplets with high-confidence SNP-gene pairs derived from multiple S2G linking strategies, we constructed a comprehensive set of 489 high-confidence {regulatory SNP-context-gene-cancer} quadruplets across nine cancers (counts by cancer in Table 3 ). Table 4 highlights illustrative quadruplets for each cancer ranked by CT-FM-SNP PIP values, displaying up to five quadruplets per cancer where available. When multiple quadruplets existed for the same SNP-gene pair across different biological contexts, we present the context with the highest PIP value. View this table: View inline View popup Download powerpoint Table 4. Examples of high-confidence regulatory quadruplets linking cancer risk variants to putative target genes and biological contexts. This analysis yielded high-confidence regulatory mechanisms for particular SNPs within tissue-of-origin contexts for overall and ER+ breast cancer (e.g., rs68056147- LINC-PINT -mammary luminal epithelial cells), prostate cancer (e.g., rs3916132- ACPP -VCaP), colorectal cancer (only one quadruplet found: rs35943851- METRNL -COAD), and renal cancer (e.g., rs11023840- KCNQ1OT1 -KIRC). For other cancers, the links between genes and contexts were less specific due to the absence of any single dominant signal—for example, bladder cancer SNP rs5750711 and its target gene APOBEC3B showed competing signals among seven contexts including PC-9 lung adenocarcinoma cell line and bladder urothelial carcinoma (BLCA) bulk tumor tissue, with CT-FM-SNP PIPs ranging from 0.51 to 0.59. Details of all identified quadruplets can be found in Table S4 . Sensitivity analyses excluding bulk tumor annotations When excluding TCGA bulk tumor annotations, CT-FM consistently identified mammary luminal epithelial cells as the high-confidence biological context for both overall breast cancer and its ER+ subtype (PIP = 0.62-0.75), and VCaP as the high-confidence context for prostate cancer (PIP = 1.00). No high-confidence contexts were identified for the remaining cancer types ( Table S5 ). In the corresponding CT-FM-SNP analysis, the most frequently detected contexts for most cancer types remained concordant with the primary analysis, except for renal cancer, where the predominant context shifted from KIRC to renal proximal tubule epithelial cells and peripheral blood naïve B cells ( Table S6 ). Complete {regulatory SNP-context-gene-cancer} quadruplets derived from this analysis are provided in Table S7 . Discussion In this study, we employed an integrative statistical fine-mapping framework to systematically elucidate the cellular origins of genetic susceptibility across 11 solid cancers. By integrating large-scale GWAS data with an atlas of approximately 1,500 context-specific cis -regulatory elements, our approach advanced beyond traditional heritability partitioning to identify potential cancer-relevant biological contexts. A central contribution of our work lies in the substantial refinement of putative causal signals achieved through statistical fine-mapping. Whereas conventional enrichment analysis (i.e., S-LDSC) implicated 141 context-specific annotations, our fine-mapping process, which explicitly models annotation co-regulation, distilled these associations into four high-confidence biological contexts for breast, prostate, colorectal, and renal cancers, respectively. This represents an advancement in specificity, enabling the disentanglement of putative causal contexts from bystander signals attributed to shared regulatory programs. Our genome-wide and variant-specific analyses consistently converged on findings that reinforce established biological paradigms in cancer etiology. At a broad level, genome-wide analyses identified tissue-of-origin epithelial contexts as the predominant drivers of heritable risk for these cancers. This pattern was exemplified by the prioritization of mammary luminal epithelial cells for both overall and ER+ breast cancer, consistent with their established role as the presumptive cells of origin for breast malignancies. 39 , 40 The strong signal for the VCaP epithelial cell line, derived from a prostate cancer vertebral metastasis, suggests that the regulatory architecture disrupted by germline risk variants is not only critical for disease initiation but may also be retained and relevant in advanced, metastatic states. 41 , 42 The prioritization of bulk tumor tissues for colorectal and renal cancers is consistent with this observation. The bulk tumor may act as a proxy for the true cell of origin, retaining the regulatory signatures shaped by germline variants through malignant transformation. 43 The signal may also arise from the collective contribution of non-cancerous cell types—such as resident immune and stromal populations—that coinhabit tumors. 8 Both interpretations point to the in vivo tissue environment as the critical context for understanding how germline variants confer cancer risk. Variant-level fine-mapping closely aligned with these results, with the same epithelial and bulk tumor contexts frequently emerging as the prioritized context for individual risk variants identified through functionally informed fine-mapping, while also providing additional insights at higher resolution. For instance, in prostate cancer, the bulk prostate tumor tissue—a candidate signal in the genome-wide analysis—emerged as the most frequent high-confidence context for individual SNPs alongside VCaP. Similarly, for ER– breast cancer, which lacked a genome-wide signal, the variant-level approach successfully validated bulk breast tumor tissue as a key etiologic context. 44 These findings demonstrate the complementary value of multi-scale analytical approaches in dissecting cancer heritability. Beyond the observations aligning with established cell– and tissue-of-origin biology, our analysis also uncovered less-obvious yet biologically plausible contexts warranting further investigation. Our results underscore the critical role of the TME; 8 for instance, the implication of various stromal cell types in colorectal cancer, while not colonic in origin themselves, likely points to a shared, pan-fibroblastic regulatory program, or reflects these annotations acting as strong statistical proxies for the true causal cell populations within the colon’s stroma. The contribution of TME may also be reflected in the implication of immune cell types, such as naïve B cells, in renal cancer, corroborating the importance of the immune response in cancer susceptibility. 8 Furthermore, our discoveries point to potential for cross-tissue etiologies, suggesting shared biological programs underlying cancer risk across different organs. Notable examples include a compelling set of signals for bladder cancer supported by both prostate epithelial cells and fetal ureteric bud cells, consistent with their common developmental origin and related epithelial lineages; 45 and the association of bulk endometrial tumor tissue (UCEC) with ER+, but not ER–, breast cancer risk, corroborating known shared hormonal signaling pathways. 46 To translate these biological contexts into more specific molecular mechanisms, we mapped our context-specific causal variants to their putative regulatory genes using a heritability-based framework combining various S2G linking strategies. 38 A review of these connections reveals that many of the mapped genes have well-documented or plausible roles in cancer biology. Examples include the established tumor-suppressive function of LINC-PINT in breast cancer 47 and oncogenic role of CLPTM1L in lung cancer, 48 as well as emerging connections such as DPF3 , a chromatin remodeling factor previously implicated in kidney cancer cell migration, 49 and TNS1 , whose role in cell invasiveness may contribute to breast cancer risk. 50 Integrating these gene links with our prior results allowed us to construct 489 {regulatory SNP-context-gene-cancer} quadruplets. The credibility of these proposed mechanisms is underscored by cases where our framework points to interactions already supported by external evidence. For example, our quadruplet for prostate cancer links genetic risk to the ACPP gene within prostate epithelial cells— the very cells where prostatic acid phosphatase (ACPP) is known to be synthesized. 51 , 52 Similarly, our link between KCNQ1OT1 and genetic susceptibility to renal cancer in bulk kidney tumor tissue is supported by the gene’s known high expression in KIRC tumors. 53 Beyond reinforcing known interactions, these quadruplets propose novel testable mechanisms, each defining a full regulatory path from variant to gene, context, and ultimately, to cancer. Interestingly, our results also illuminate potential mechanisms of pleiotropy, revealing how loci near TERT , TET2 , FGFR2 , CDKN1A , RHOD , and ANXA9 are implicated across multiple malignancies. For example, the RHOD locus (11q13.2) is implicated in both ER+ breast cancer (mammary luminal epithelial cells) and prostate cancer (VCaP and PRAD) specifically through variant rs114895445, and the CDKN1A locus (6p21.2) is linked to both prostate (VCaP) and renal cancer (KIRC) through variant rs762624 (illustrated as an example in Figure 3 ), highlighting cancer-shared genetic pathways that may be activated through distinct biological contexts. Download figure Open in new tab Figure 3. Context-specific pleiotropy at CDKN1A locus in prostate and renal cancer. ( A , B ) Regional association plots for prostate and renal cancer at CDKN1A locus (6p21.2), showing genetic variant associations within a 50 kb window surrounding the putatively causal variant rs762624 (purple diamond; PolyFun+SuSiE posterior inclusion probability [PIP] ≥ 0.5). Light blue shaded regions delineate CDKN1A gene boundaries (chr6:36,644,240-36,655,109, GRCh37/hg19; NCBI Gene). Y-axes show the P -value of association on a –log 10 scale. ( C , D ) The top five biological contexts for rs762624 as prioritized by CT-FM-SNP analysis in prostate and renal cancer, respectively, and ranked by PIP. Several limitations warrant consideration when interpreting our findings. Our analytical power is inherently dependent on the quality and scope of its inputs: specifically, the statistical strength of the underlying GWAS, the external LD reference panel, and the comprehensiveness of our functional annotation library. The dependency on well-powered GWAS likely explains the limited results for cancers with smaller sample sizes, such as esophageal cancer (n=4,112 cases), and consequently constrained our study’s scope to these 11 solid cancers, precluding analysis of other malignancies for which large-scale summary statistics are not yet available (or publicly accessible). We were also limited by the lack of availability of large-scale GWAS from non-European ancestry populations. Future investigations can apply this framework in a multi-ancestry setting, harnessing shared genetic effects and diverse LD patterns across populations to enhance statistical power and fine-mapping resolution. 54 Furthermore, if the truly causal cell type or precise biological context is not included in our reference panel, our context fine-mapping model may identify the best available proxy cell types or tissues. Therefore, an identified context should be interpreted as representing a broader relevant biological program rather than a literal, definitive cellular source. This same principle applies to the fine-mapped variants themselves, as the true causal SNP may be absent from our analysis due to initial SNP reference panel limitations or filtering. Consequently, all findings presented here are data-driven hypotheses that necessitate at least two crucial avenues of follow-up: first, replication and expansion in larger, more diverse GWAS cohorts and annotation datasets —including single-cell data from relevant tumor types to deconvolve bulk tissue signals—and second, extensive laboratory functional validation to confirm their biological mechanisms before any etiological relevance can be considered. Conclusion To conclude, we applied an integrative statistical fine-mapping framework that leverages cis -regulatory element annotations to prioritize context-specific genetic risk mechanisms across 11 solid cancers. By progressing systematically from genome-wide patterns to single-variant effects, our analysis disentangled putative causal contexts from correlated signals, revealing both established and novel regulatory mechanisms. These findings advance our understanding of the genetic susceptibility to different cancers and provide a potential roadmap to guide experimental studies aimed at translating genetic discoveries into mechanistic and, ultimately, clinical insights. Funding S.L. was supported by U01CA194393. TO’M is supported by a National Health and Medical Research Council (NHMRC) of Australia Investigator Fellowship (APP1173170). This research was supported in part by the Intramural Research Program of the National Institutes of Health (NIH). The contributions of the NIH authors are considered Works of the United States Government. The findings and conclusions presented in this paper are those of the authors and do not necessarily reflect the views of the NIH or the U.S. Department of Health and Human Services. Competing interest The authors declare that they have no competing interests. Data availability All publicly available GWAS summary statistics following standardized quality control procedures have been uploaded to https://zenodo.org/records/15676245 , except for the bladder cancer dataset, which has controlled access through dbGAP phs003342.v1.p1 for research related to bladder cancer. Processing procedures and detailed data characteristics are also documented at https://github.com/xueyaowunci/cancer-context-fine-mapping . Analysis scripts and source of all cis-regulatory element annotations can be found at https://github.com/ArtemKimUSC/CTFM . Authors’ contributions XW and PK designed the study. XW collected and processed all GWAS summary statistics with the help of SL. AK curated the CT-FM software and provided all context-specific annotation data. XW performed the primary analyses with the assistance of AK, PK, and NM. XW, PK, CEB, SL, AK, NM, TOM, DR, TD, SK, NR, and LPO interpreted the results. XW drafted the manuscript with significant input from PK. All co-authors contributed to review and revision of the manuscript. Ethics declaration This study utilized exclusively summary-level GWAS data from previous studies. No individual-level genetic or phenotypic data were accessed or analyzed. As all data used were aggregate summary statistics that do not contain identifiable information about individual participants, this research did not require institutional review board approval or ethics committee oversight. The original studies from which the summary statistics were derived had obtained appropriate ethical approvals and participant consent as reported in their respective publications. Acknowledgments We gratefully acknowledge all researchers and consortia who have made GWAS summary statistics and regulatory annotations publicly available or shared data through collaborative arrangements. The willingness of the scientific community to share these valuable datasets and annotations is essential for advancing our understanding of complex trait genetics and regulatory mechanisms, and enables secondary analyses such as this study. We thank the original study participants and investigators whose contributions made these datasets possible. References 1. ↵ Bray , F. et al. Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries . CA Cancer J Clin 74 , 229 – 263 ( 2024 ). OpenUrl CrossRef PubMed 2. ↵ Bizuayehu , H. M. et al. Global Disparities of Cancer and Its Projected Burden in 2050 . JAMA Netw Open 7 , e2443198 – e2443198 ( 2024 ). OpenUrl CrossRef 3. ↵ Sur , I. , Tuupanen , S. , Whitington , T. , Aaltonen , L. A. & Taipale , J . Lessons from functional analysis of genome-wide association studies . Cancer Res 73 , 4180 – 4184 ( 2013 ). OpenUrl Abstract / FREE Full Text 4. ↵ Schipper , M. & Posthuma , D . Demystifying non-coding GWAS variants: an overview of computational tools and methods . Hum Mol Genet 31 , R73 ( 2022 ). OpenUrl CrossRef PubMed 5. ↵ Wang , Y. & Hon , G. C . Towards functional maps of non-coding variants in cancer . Front Genome Ed 6 , 1481443 ( 2024 ). OpenUrl CrossRef PubMed 6. ↵ Kim-Hellmuth , S. et al. Cell type–specific genetic regulation of gene expression across human tissues . Science (1979) 369 , ( 2020 ). 7. ↵ Hekselman , I. & Yeger-Lotem , E . Mechanisms of tissue and cell-type specificity in heritable traits and diseases . Nature Reviews Genetics 2020 21:3 21 , 137 – 150 ( 2020 ). OpenUrl PubMed 8. ↵ de Visser , K. E. & Joyce , J. A . The evolving tumor microenvironment: From cancer initiation to metastatic outgrowth . Cancer Cell 41 , 374 – 403 ( 2023 ). OpenUrl CrossRef PubMed 9. ↵ Meyer , K. B. et al. Fine-scale mapping of the FGFR2 breast cancer risk locus: Putative functional variants differentially bind FOXA1 and E2F1 . Am J Hum Genet 93 , 1046 – 1060 ( 2013 ). OpenUrl CrossRef PubMed 10. ↵ Long , E. et al. Context-aware single-cell multiomics approach identifies cell-type-specific lung cancer susceptibility genes . Nature Communications 2024 15:1 15 , 1 – 16 ( 2024 ). OpenUrl PubMed 11. ↵ Hekselman , I. & Yeger-Lotem , E . Mechanisms of tissue and cell-type specificity in heritable traits and diseases . Nat Rev Genet 21 , 137 – 150 ( 2020 ). OpenUrl PubMed 12. ↵ Zhang , J. et al. An integrative ENCODE resource for cancer genomics . Nat Commun 11 , 3696 ( 2020 ). OpenUrl CrossRef PubMed 13. ↵ Corces , M. R. et al. The chromatin accessibility landscape of primary human cancers . Science (1979) 362 , ( 2018 ). 14. ↵ Finucane , H. K. et al. Partitioning heritability by functional annotation using genome-wide association summary statistics . Nat Genet 47 , 1228 – 1235 ( 2015 ). OpenUrl CrossRef PubMed 15. ↵ Finucane , H. K. et al. Heritability enrichment of specifically expressed genes identifies disease-relevant tissues and cell types . Nat Genet 50 , 621 – 629 ( 2018 ). OpenUrl CrossRef PubMed 16. ↵ Ren , Y. Y. & Liu , Z . Characterization of Single-Cell Cis-regulatory Elements Informs Implications for Cell Differentiation . Genome Biol Evol 16 , ( 2024 ). 17. ↵ Sheffield , N. C. et al. Patterns of regulatory activity across diverse human cell types predict tissue identity, transcription factor binding, and long-range interactions . Genome Res 23 , 777 ( 2013 ). OpenUrl Abstract / FREE Full Text 18. ↵ Gazal , S. et al. Linkage disequilibrium dependent architecture of human complex traits shows action of negative selection . Nat Genet 49 , 1421 ( 2017 ). OpenUrl CrossRef PubMed 19. ↵ Amariuta , T. , Siewert-Rocks , K. & Price , A. L . Modeling tissue co-regulation estimates tissue-specific contributions to disease . Nat Genet 55 , 1503 – 1511 ( 2023 ). OpenUrl CrossRef PubMed 20. ↵ Kim , A. et al. Inferring causal cell types of human diseases and risk variants from candidate regulatory elements . medRxiv ( 2024 ) doi: 10.1101/2024.05.17.24307556 . OpenUrl Abstract / FREE Full Text 21. ↵ Weissbrod , O. et al. Functionally informed fine-mapping and polygenic localization of complex trait heritability . Nat Genet 52 , 1355 – 1363 ( 2020 ). OpenUrl CrossRef PubMed 22. ↵ SCREEN : Search Candidate cis-Regulatory Elements by ENCODE . 23. ↵ Zhang , K. et al. A single-cell atlas of chromatin accessibility in the human genome . Cell 184 , 5985 – 6001 .e19 ( 2021 ). OpenUrl CrossRef PubMed 24. ↵ Nasser , J. et al. Genome-wide enhancer maps link risk variants to disease genes . Nature 2021 593:7858 593 , 238 – 243 ( 2021 ). OpenUrl CrossRef PubMed 25. ↵ Boix , C. A. , James , B. T. , Park , Y. P. , Meuleman , W. & Kellis , M . Regulatory genomic circuitry of human disease loci by integrative epigenomics . Nature 2021 590:7845 590 , 300 – 307 ( 2021 ). OpenUrl CrossRef PubMed 26. ↵ Michailidou , K. et al. Association analysis identifies 65 new breast cancer risk loci . Nature 551 , 92 – 94 ( 2017 ). OpenUrl CrossRef PubMed 27. Phelan , C. M. et al. Identification of 12 new susceptibility loci for different histotypes of epithelial ovarian cancer . Nat Genet 49 , 680 – 691 ( 2017 ). OpenUrl CrossRef PubMed 28. Wang , A. et al. Characterizing prostate cancer risk through multi-ancestry genome-wide discovery of 187 novel risk variants . Nat Genet 55 , 2065 – 2074 ( 2023 ). OpenUrl CrossRef PubMed 29. Fernandez-Rozadilla , C. et al. Deciphering colorectal cancer genetics through multi-omic analysis of 100,204 cases and 154,587 controls of European and east Asian ancestries . Nat Genet 55 , 89 – 99 ( 2023 ). OpenUrl CrossRef PubMed 30. Gharahkhani , P. et al. Genome-wide association studies in oesophageal adenocarcinoma and Barrett’s oesophagus: a large-scale meta-analysis . Lancet Oncol 17 , 1363 – 1373 ( 2016 ). OpenUrl CrossRef PubMed 31. McKay , J. D. et al. Large-scale association analysis identifies new lung cancer susceptibility loci and heterogeneity in genetic susceptibility across histological subtypes . Nat Genet 49 , 1126 – 1132 ( 2017 ). OpenUrl CrossRef PubMed 32. Purdue , M. P. et al. Multi-ancestry genome-wide association study of kidney cancer identifies 63 susceptibility regions . Nat Genet 56 , 809 – 818 ( 2024 ). OpenUrl PubMed 33. Koutros , S. et al. Genome-wide Association Study of Bladder Cancer Reveals New Biological and Translational Insights . Eur Urol 84 , 127 – 137 ( 2023 ). OpenUrl PubMed 34. ↵ Ramachandran , D. et al. GWAS meta-analysis identifies five susceptibility loci for endometrial cancer . EBioMedicine 118 , 105830 ( 2025 ). 35. ↵ Zhang , H. et al. Genome-wide association study identifies 32 novel breast cancer susceptibility loci from overall and subtype-specific analyses . Nat Genet 52 , 572 – 581 ( 2020 ). OpenUrl CrossRef PubMed 36. ↵ Wang , G. , Sarkar , A. , Carbonetto , P. & Stephens , M . A simple new approach to variable selection in regression, with application to genetic fine mapping . J R Stat Soc Series B Stat Methodol 82 , 1273 – 1300 ( 2020 ). OpenUrl CrossRef PubMed 37. ↵ Zou , Y. , Carbonetto , P. , Wang , G. & Stephens , M . Fine-mapping from summary data with the “Sum of Single Effects” model . PLoS Genet 18 , e1010299 ( 2022 ). OpenUrl CrossRef PubMed 38. ↵ Gazal , S. et al. Combining SNP-to-gene linking strategies to identify disease genes and assess disease omnigenicity . Nat Genet 54 , 827 – 836 ( 2022 ). OpenUrl CrossRef PubMed 39. ↵ Sayaman , R. W. et al. Luminal epithelial cells integrate variable responses to aging into stereotypical changes that underlie breast cancer susceptibility . Elife 13 , ( 2024 ). 40. ↵ Hein , S. M. et al. Luminal Epithelial Cells within the Mammary Gland Can Produce Basal Cells upon Oncogenic Stress . Oncogene 35 , 1461 ( 2015 ). OpenUrl CrossRef PubMed 41. ↵ Chatrath , A. , Ratan , A. & Dutta , A . Germline Variants that Affect Tumor Progression . Trends Genet 37 , 433 ( 2020 ). OpenUrl PubMed 42. ↵ Quigley , D. A. et al. Genomic Hallmarks and Structural Variation in Metastatic Prostate Cancer . Cell 174 , 758 – 769 .e9 ( 2018 ). OpenUrl CrossRef PubMed 43. ↵ Hoadley , K. A. et al. Cell-of-Origin Patterns Dominate the Molecular Classification of 10,000 Tumors from 33 Types of Cancer . Cell 173 , 291 – 304 .e6 ( 2018 ). OpenUrl CrossRef PubMed 44. ↵ Thennavan , A. et al. Molecular analysis of TCGA breast cancer histologic types . Cell Genomics 1 , 100067 ( 2021 ). OpenUrl PubMed 45. ↵ Cunha , G. R. et al. Development of the human prostate . Differentiation 103 , 24 ( 2018 ). OpenUrl CrossRef PubMed 46. ↵ Low , Y. L. et al. Multi-Variant Pathway Association Analysis Reveals the Importance of Genetic Determinants of Estrogen Metabolism in Breast and Endometrial Cancer Susceptibility . PLoS Genet 6 , e1001012 ( 2010 ). OpenUrl CrossRef PubMed 47. ↵ Li , D. & Hu , A . LINC-PINT suppresses breast cancer cell proliferation and migration via MEIS2/PPP3CC/NF-κB pathway by sponging miR-576-5p . American Journal of the Medical Sciences 367 , 201 – 211 ( 2024 ). OpenUrl PubMed 48. ↵ James , M. A. , Vikis , H. G. , Tate , E. , Rymaszewski , A. L. & You , M . CRR9/CLPTM1L Regulates Cell Survival Signaling and is Required for Ras Transformation and Lung Tumorigenesis . Cancer Res 74 , 1116 ( 2013 ). OpenUrl PubMed 49. ↵ Cui , H. et al. The SWI/SNF chromatin remodeling factor DPF3 regulates metastasis of ccRCC by modulating TGF-β signaling . Nat Commun 13 , 4680 ( 2022 ). OpenUrl PubMed 50. ↵ Wang , Z. et al. TNS1: Emerging Insights into Its Domain Function , Biological Roles, and Tumors. Biology (Basel ) 11 , 1571 ( 2022 ). OpenUrl PubMed 51. ↵ Li , S. S. L. & Sharief , F. S . The prostatic acid phosphatase (acpp) gene is localized to human chromosome 3q21-q23 . Genomics 17 , 765 – 766 ( 1993 ). OpenUrl PubMed 52. ↵ Aguet , F. et al. The GTEx Consortium atlas of genetic regulatory effects across human tissues . Science (1979) 369 , 1318 – 1330 ( 2020 ). OpenUrl Abstract / FREE Full Text 53. ↵ Lin , Z. Bin et al. Long Noncoding RNA KCNQ1OT1 is a Prognostic Biomarker and mediates CD8+ T cell exhaustion by regulating CD155 Expression in Colorectal Cancer . Int J Biol Sci 17 , 1757 ( 2021 ). OpenUrl PubMed 54. ↵ Yuan , K. et al. Fine-mapping across diverse ancestries drives the discovery of putative causal variants underlying human complex traits and diseases . Nat Genet 56 , 1841 – 1850 ( 2024 ). OpenUrl CrossRef PubMed View the discussion thread. Back to top Previous Next Posted November 02, 2025. Download PDF Supplementary Material Email Thank you for your interest in spreading the word about medRxiv. NOTE: Your email address is requested solely to identify you as the sender of this article. Your Email * Your Name * Send To * Enter multiple addresses on separate lines or separate them with commas. You are going to email the following Prioritizing context-specific genetic risk mechanisms in 11 solid cancers Message Subject (Your Name) has forwarded a page to you from medRxiv Message Body (Your Name) thought you would like to see this page from the medRxiv website. Your Personal Message CAPTCHA This question is for testing whether or not you are a human visitor and to prevent automated spam submissions. Share Prioritizing context-specific genetic risk mechanisms in 11 solid cancers Xueyao Wu , Artem Kim , Charles E Breeze , Tracy A O’Mara , Dhanya Ramachandran , Thilo Dörk , Stella Koutros , Nathaniel Rothman , Ludmila Prokunina-Olsson , Nicholas Mancuso , Sara Lindström , Peter Kraft medRxiv 2025.10.30.25339145; doi: https://doi.org/10.1101/2025.10.30.25339145 Share This Article: Copy Citation Tools Prioritizing context-specific genetic risk mechanisms in 11 solid cancers Xueyao Wu , Artem Kim , Charles E Breeze , Tracy A O’Mara , Dhanya Ramachandran , Thilo Dörk , Stella Koutros , Nathaniel Rothman , Ludmila Prokunina-Olsson , Nicholas Mancuso , Sara Lindström , Peter Kraft medRxiv 2025.10.30.25339145; doi: https://doi.org/10.1101/2025.10.30.25339145 Citation Manager Formats BibTeX Bookends EasyBib EndNote (tagged) EndNote 8 (xml) Medlars Mendeley Papers RefWorks Tagged Ref Manager RIS Zotero Tweet Widget Facebook Like Google Plus One Subject Area Epidemiology Subject Areas All Articles Addiction Medicine (567) Allergy and Immunology (863) Anesthesia (297) Cardiovascular Medicine (4411) Dentistry and Oral Medicine (443) Dermatology (380) Emergency Medicine (606) Endocrinology (including Diabetes Mellitus and Metabolic Disease) (1505) Epidemiology (15205) Forensic Medicine (30) Gastroenterology (1119) Genetic and Genomic Medicine (6574) Geriatric Medicine (666) Health Economics (994) Health Informatics (4511) Health Policy (1365) Health Systems and Quality Improvement (1608) Hematology (537) HIV/AIDS (1263) Infectious Diseases (except HIV/AIDS) (15903) Intensive Care and Critical Care Medicine (1103) Medical Education (620) Medical Ethics (144) Nephrology (665) Neurology (6573) Nursing (345) Nutrition (998) Obstetrics and Gynecology (1139) Occupational and Environmental Health (954) Oncology (3319) Ophthalmology (967) Orthopedics (369) Otolaryngology (420) Pain Medicine (435) Palliative Medicine (129) Pathology (662) Pediatrics (1689) Pharmacology and Therapeutics (691) Primary Care Research (710) Psychiatry and Clinical Psychology (5421) Public and Global Health (9205) Radiology and Imaging (2191) Rehabilitation Medicine and Physical Therapy (1367) Respiratory Medicine (1191) Rheumatology (593) Sexual and Reproductive Health (709) Sports Medicine (529) Surgery (709) Toxicology (99) Transplantation (288) Urology (265) (function(){function c(){var b=a.contentDocument||a.contentWindow.document;if(b){var d=b.createElement('script');d.innerHTML="window.__CF$cv$params={r:'9fe8c4554b5241e2',t:'MTc3OTI1MzExMw=='};var a=document.createElement('script');a.src='/cdn-cgi/challenge-platform/scripts/jsd/main.js';document.getElementsByTagName('head')[0].appendChild(a);";b.getElementsByTagName('head')[0].appendChild(d)}}if(document.body){var a=document.createElement('iframe');a.height=1;a.width=1;a.style.position='absolute';a.style.top=0;a.style.left=0;a.style.border='none';a.style.visibility='hidden';document.body.appendChild(a);if('loading'!==document.readyState)c();else if(window.addEventListener)document.addEventListener('DOMContentLoaded',c);else{var e=document.onreadystatechange||function(){};document.onreadystatechange=function(b){e(b);'loading'!==document.readyState&&(document.onreadystatechange=e,c())}}}})();