LYN Mutations in Breast Cancer: Association with Central Nervous System Metastasis and Domain-Level Insights

LYN Mutations in Breast Cancer: Association with Central Nervous System Metastasis and Domain-Level Insights | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (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],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article LYN Mutations in Breast Cancer: Association with Central Nervous System Metastasis and Domain-Level Insights Elif Kardelen Çağdaş, Berkay Çağdaş This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7597855/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 05 Dec, 2025 Read the published version in BMC Cancer → Version 1 posted 12 You are reading this latest preprint version Abstract Objectives: Central nervous system (CNS) metastasis is a major driver of morbidity in metastatic breast cancer, yet the molecular determinants of CNS tropism remain incompletely defined. LYN , a Src-family kinase integrating receptor tyrosine kinase and integrin signaling, is a biologically plausible mediator of metastatic traits. Design: We performed a retrospective, multi-study analysis of publicly available breast cancer cohorts aggregated in cBioPortal. After harmonization and de-duplication, LYN status was determinable in 5,947 invasive carcinoma of no special type (NST) tumors across 29 studies. The primary endpoint was CNS metastasis at any time (Yes/No), harmonized via a prespecified controlled vocabulary (case-insensitive substring mapping). Somatic LYN variants (coding SNVs/indels) were collapsed to patient-level classes (missense-only; truncating if any nonsense/frameshift/splice). Variants with resolvable positions were mapped to Src-family modules (SH4/Unique, SH3, SH2, SH2–kinase linker, kinase). Two-group comparisons used two-sided Fisher’s exact tests with exact 95% CIs; domain screens used omnibus χ² and Benjamini–Hochberg FDR control. A prespecified Firth logistic model evaluated truncating vs missense within LYN -mutant tumors. Setting: Public cancer genomics repositories (cBioPortal); multi-institutional cohorts. Participants: 5,947 tumors across multi-study cohorts with LYN status available. Interventions: None. Main outcome measures: Primary: ever-CNS metastasis (yes/no). Secondary: distribution of LYN variant classes and domains (SH4/Unique, SH3, SH2, linker, kinase). Results: CNS metastasis occurred in 5/46 (10.9%) LYN -mutated tumors vs 110/5,901 (1.9%) LYN wild-type tumors (OR = 6.42; 95% CI, 2.49–16.56; p = 0.0018). Within LYN -mutant cases, domain distributions differed by CNS status (omnibus χ² p ≈ 0.014); a one-versus-rest signal at the SH4/Unique N-terminus was nominally significant and borderline after FDR (unadjusted p ≈ 0.010; q ≈ 0.052; small in-domain n = 3). By mutation class, truncating vs missense showed a higher CNS-positive proportion (28.6% vs 7.9%) but did not reach significance (Fisher p = 0.166; alternatively framed OR = 4.22; exact 95% CI, 0.58–30.75; p = 0.182). Firth estimates were directionally consistent with wide profile CIs under sparse counts. Conclusions: Across pooled cohorts, LYN mutation is associated with increased odds of CNS metastasis, and domain context appears informative, with a small-sample, FDR-borderline enrichment at the SH4/Unique N-terminus. The truncating-class signal is exploratory given limited power. These data prioritize domain-aware LYN annotation for independent validation and mechanistic follow-up focused on membrane targeting, endothelial adhesion, and blood–brain barrier traversal. Trial registration: Not applicable. breast cancer cBioPortal CNS metastasis domain mapping Fisher’s exact Firth logistic LYN SH4/Unique Src-family kinase Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Key messages What is already known: CNS metastasis is a major clinical problem in breast cancer; molecular determinants of brain tropism remain incompletely defined. What this study adds: Across pooled cohorts, any LYN mutation is associated with higher odds of CNS metastasis; domain context (notably SH4/Unique) may matter, albeit with small counts and FDR-borderline evidence. How this might affect research/practice/policy: Encourages domain-aware LYN annotation and targeted mechanistic work on membrane targeting/BBB traversal; not ready for risk stratification or treatment selection. 1. Introduction Breast cancer remains a major cause of cancer morbidity and mortality, with a substantial share driven by CNS involvement ( 1 ). As systemic therapies extend survival, brain and leptomeningeal metastases have greater clinical impact, yet the molecular drivers of CNS tropism are still poorly defined ( 2 ). Brain colonization demands tumor-intrinsic programs—motility, invasion, survival—plus BBB transmigration and adaptation to the neural microenvironment ( 3 ). Signaling modules that couple cell adhesion to RTK inputs and cytoskeletal regulation are thus strong candidates ( 4 ). Src family kinases—especially LYN —sit at this nexus, integrating integrin/FAK signaling, ERBB/EGFR cross-talk, and pathways controlling EMT, cell–cell junctions, and actin remodeling ( 5 ). Although LYN copy-number/expression changes recur across cancers, the clinical relevance of specific LYN variants in breast cancer, particularly for CNS involvement, remains undefined ( 6 ), motivating a focused analysis of LYN ’s structure–function context and an exploratory test of whether mutation class and domain location associate with CNS tropism. 1.1. LYN Within the Src Family: Overview and Relevance to Breast Cancer LYN is a non-receptor tyrosine kinase in the Src family (SFKs), conserved signal integrators that relay inputs from cell-surface receptors to control proliferation, survival, adhesion, motility, and cytoskeletal dynamics ( 7 ). SFKs are rapidly engaged downstream of RTKs, integrins, GPCRs, immune and cytokine receptors, and phosphorylate substrates that scaffold focal adhesions, remodel actin, and amplify mitogenic/pro-survival cascades ( 5 ). Because of this nodal position, small shifts in SFK activity or localization can produce outsized cellular effects. In immunity, LYN frequently acts as a negative regulator yet can also activate signaling, being required for both initiation and termination of B-cell responses (Fig. 1 ). Physiologically, LYN tunes B-cell receptor signaling, Fc pathways in myeloid/mast cells, and tonic thresholds for activation/tolerance, acting positively or negatively depending on the complex and adaptors ( 8 ). Upstream inputs include BCR (CD79A/B), CD5, CD19, CD22, Fc receptors, TLR2/4, and receptors such as EPOR, KIT, MPL, CXCR4, and IL-3/IL-5/CSF2 ( 9 ). Outside hematopoiesis, LYN in epithelial tissues—including breast cancer—regulates cell–matrix adhesion, migration, EMT, and, via integrin/FAK and RTK (ERBB/EGFR) coupling to PI3K–AKT and RAS–RAF–MEK–ERK, enhances survival and proliferation ( 9 , 10 ). It also dampens signaling by phosphorylating ITIMs to recruit SHP-1/2 and SHIP-1. LYN ’s qualitative output depends on conformation, localization, and scaffold/phosphatase balance, explaining context-dependent effects ( 11 ). In breast cancer, altered LYN expression/mutation and rewired RTK–integrin–EMT networks plausibly promote CNS tropism—from intravasation and circulatory survival to endothelial adhesion and BBB transmigration ( 6 , 12 ). 1.2. Molecular Architecture of LYN Like other Src family kinases (SFKs), LYN comprises modular domains that govern membrane targeting, autoinhibition, and stimulus-coupled activation: N-terminus / SH4-Unique: An N-terminal glycine is myristoylated to enable membrane association; palmitoylation and basic residues in the Unique region further stabilize interactions with specific membrane microdomains and organelles. This region dictates subcellular localization-and thus receptor/scaffold access-and alternative splicing fine-tunes these effects ( 5 ). SH3 domain: Binds proline-rich motifs in partners and in LYN 's own linker, clamping the kinase in a closed, autoinhibited state. High-affinity exogenous ligands can outcompete the intramolecular interaction to promote opening and priming ( 13 ). SH2 domain: Recognizes phosphotyrosine motifs. In autoinhibition, it engages the phosphorylated C-terminal regulatory tyrosine, reinforcing the closed conformation; binding to external pTyr ligands displaces this contact, activates the kinase, and recruits LYN to new complexes ( 13 ). SH2-kinase linker: A flexible, proline-rich segment that transmits conformational signals from the SH3/SH2 "regulatory head" to the catalytic core, acting as a mechanical gate that shifts the closed-open equilibrium and modulates catalytic efficiency ( 13 ). Kinase domain: The bilobal catalytic core contains the activation loop, whose phosphorylation state and conformation control substrate access and turnover. Subtle changes in loop dynamics-via mutation or scaffolding-can markedly alter activity and specificity ( 14 ). C-terminal regulatory tyrosine / CSK axis: CSK-mediated phosphorylation of the tail tyrosine promotes SH2 docking and autoinhibition; dephosphorylation by receptor-proximal phosphatases releases the clamp. The CSK-phosphatase balance thus sets a tunable rheostat on LYN activity ( 15 ). Collectively, these modules create a conformational switch that integrates lipid targeting, intramolecular clamps, and catalytic licensing to couple localization to signaling (Fig. 2 ). In epithelial and breast cancer settings, perturbations in membrane targeting, regulatory interfaces, or the kinase core can yield distinct phenotypes-altered focal-adhesion dynamics, EMT programs, and RTK/integrin responses-motivating tests of whether variant classes (e.g., truncations vs. interface/targeting missense changes) associate with CNS metastasis ( 6 , 13 – 15 ). 1.3. LYN Alterations in Breast Cancer Copy-number and expression changes: Across cancers, amplification/overexpression of LYN has been reported more frequently than recurrent hotspot mutations, suggesting dosage-sensitive contributions to signaling output and plasticity ( 9 ). In breast cancer, point mutations are relatively rare and heterogeneous in position and type (missense, occasional truncations/splice). Most single-nucleotide variants lack definitive functional annotation, and their effects likely depend on domain context (e.g., SH4/Unique vs kinase core) and cellular compartmentalization ( 16 ). For the majority of variants, oncogenicity remains uncertain and there is no established clinical actionability. Consequently, mapping variants to structural/functional modules and to phenotypes such as CNS metastasis may help prioritize variants for mechanistic study and clinical surveillance. Despite extensive pathway crosstalk with receptor tyrosine kinases and integrins, there are no approved LYN -directed therapies in breast cancer; early clinical experience with multi-target Src family kinase (SFK) inhibitors has been mixed, highlighting the importance of biomarker-guided strategies rather than unselected use ( 17 ). 1.4. CNS Metastasis Biology and the Blood–Brain Barrier (BBB) Brain metastasis proceeds through sequential bottlenecks: (i) local invasion and intravasation, (ii) survival in circulation and vascular arrest within the cerebral microvasculature, (iii) adhesion to brain endothelium and transmigration across the BBB (paracellular and/or transcellular routes), and (iv) adaptation to the neural niche, including crosstalk with astrocytes, microglia, and perivascular cells that governs early survival and outgrowth ( 18 ). The BBB’s tight junctions, low transcytosis, and specialized perivascular architecture (endothelial cells, pericytes, astrocyte end-feet) impose stringent barrier and signaling constraints on circulating tumor cells ( 19 ). LYN could influence CNS tropism at multiple nodes: On the tumor cell, by coordinating integrin–FAK–SFK signaling for endothelial adhesion, cytoskeletal rearrangement, and diapedesis; by modulating RTK-driven survival during transit; and by altering membrane targeting (via the SH4/Unique region) that dictates receptor proximity. In the microenvironment, LYN -dependent immune signaling may shape endothelial permeability and perivascular inflammation, indirectly affecting BBB traversal. Within the endothelium, SFK activity contributes to junctional dynamics; perturbations in LYN -centered pathways could tilt the balance toward passage or retention. 1.5. Rationale, Objective, and Prespecified Hypotheses Rationale. Given LYN ’s membrane targeting, conformational regulation, and positioning at the crossroads of integrin/FAK and RTK signaling, perturbations in LYN —especially those altering regulatory interfaces or the catalytic core—could modulate key steps of CNS tropism, including endothelial adhesion and BBB transmigration. This study aims to investigate whether the class of LYN mutations (truncating vs. missense) and their domain localization are associated with CNS metastasis in breast cancer. Based on the structural logic of LYN , we hypothesize that truncating mutations—by disrupting regulatory/tail elements—and N-terminal SH4/Unique alterations—by modifying membrane targeting—are enriched in CNS-positive cases, consistent with mechanisms facilitating endothelial adhesion, BBB transmigration, and neural niche adaptation. This work provides a domain-resolved view of somatic LYN alterations in relation to CNS involvement in breast cancer, linking structural modules to metastatic phenotypes. By pooling harmonized, multi-cohort data and prespecifying LYN domain- and variant class–based hypotheses, we aim to generate mechanistically grounded, testable leads that prioritize biomarker development and guide functional validation. 2. Methods 2.1. Study Design and Cohort We conducted a retrospective analysis of publicly available breast cancer cohorts accessed via cBioPortal (20-22) (download month: September 2025). All included data are de‑identified and publicly released by the contributing studies; no direct interaction with human subjects occurred. We conducted a pooled, retrospective cohort study by harmonizing breast cancer genomics and clinical annotations across 29 independent studies, yielding an analytic cohort of 5,947 unique tumors. Harmonization steps included: (i) de-duplication using stable patient/sample identifiers across portals, (ii) standardization of variant formats and clinical fields, and (iii) exclusion of records lacking essential identifiers or LYN coverage. Inclusion criteria were histologically confirmed breast cancer, next-generation sequencing with ascertainable LYN status (wild-type vs mutant), and availability of metastatic site fields. Reporting follows STROBE recommendations for observational studies. Exclusion criteria: duplicate or discordant records, samples failing basic quality checks, or missing essential identifiers. Ethical conduct followed local regulations and the Declaration of Helsinki. Reporting adheres to STROBE recommendations for observational studies. This process encompassed 33 studies totaling n = 17,605 tumors; 29 contributed NST tumors with LYN coverage to the analytic set (n=5,947). Of these, n = 16,881 had somatic mutation profiling; applying the NST diagnosis restriction yielded n = 14,050. Within the NST set, n = 5,947 tumors (from 29 studies) had LYN sequenced and formed the analytic denominator for LYN ‑status analyses. Where a study contributed multiple samples per patient, we prioritized the earliest profiled metastatic or primary specimen per patient identifier when feasible (study‑specific rules documented in the analysis notebook). Cross‑study duplicate patients were assessed by hashed combinations of sex/age/diagnosis year and removed when confidently identified. Study-level contributions were: • breast_msk_2025 (n=1,566) • brca_tcga (n=980) • breast_ink4_msk_2021 (n=639) • breast_msk_2018 (n=415) • brca_msk_2025 (n=404) • brca_mbcproject_2022 (n=379) • brca_igr_2015 (n=216) • brca_smc_2018 (n=186) • brca_aurora_2023 (n=134) • breast_cptac_gdc (n=124) • brca_cptac_2020 (n=122) • brca_bccrc_xenograft_2014 (n=116) • brca_broad (n=103) • brca_sanger (n=100) • brca_fuscc_2020 (n=69) • brca_bccrc (n=65) • brca_mapk_hp_msk_2021 (n=61) • brca_mbcproject_wagle_2017 (n=61) • brca_pareja_msk_2020 (n=60) • brca_dfci_2020 (n=59) • bfn_duke_nus_2015 (n=22) • mbc_msk_2021 (n=18) • ilc_msk_2023 (n=14) • acbc_mskcc_2015 (n=12) • breast_alpelisib_2020 (n=10) • brca_jup_msk_2020 (n=5) • brca_hta9_htan_2022 (n=5) • brca_tcga_gdc (n=2) • brca_tcga_pan_can_atlas_2018 (n=1). Direct study links are provided in the Declarations (Availability of data and materials). Molecular subtype and missing data policy. ER/PR/HER2 molecular subtype variables were not uniformly captured across contributing studies; HER2 status in particular was missing in the majority of tumors and often inconsistently reported between primary and metastatic specimens. We therefore prespecified that subtype variables would not be included as covariates in the primary pooled analyses, and no imputation would be attempted. Because conditioning on a partially observed determinant of CNS risk (e.g., HER2) can introduce selection bias, we also refrained from restricting the analysis to complete-case subsets. 2.2. Endpoints and Variable Definitions The primary endpoint was central nervous system (CNS) metastasis at any time (yes/no), harmonized across studies using a controlled vocabulary. Metastatic sites were harmonized by mapping free-text or coded fields to a controlled vocabulary. We considered entries such as “brain”, “cerebellum”, “dura”, “occipital”, and “leptomeningeal/CSF” as CNS-positive (case-insensitive substring matching). Site fields absent were retained as missing; no imputation was performed. Dates of CNS involvement were not systematically captured across contributing studies; therefore, baseline versus incident CNS events could not be distinguished. Time-to-event analyses (e.g., Cox or Fine–Gray) were not feasible, and we prespecified a binary “ever-CNS” endpoint without imputation of missing dates. Reported associations thus reflect odds of ever having CNS involvement in the pooled dataset rather than hazards or cumulative incidence over time. Imaging intensity and follow-up schedules likely varied across studies; we did not adjust for follow-up time because event dates were unavailable. 2.3. Variant Ascertainment and Patient-Level Classification Somatic LYN variants (coding SNVs/indels) were collected from each study’s mutation tables, standardized, and collapsed to patient-level classes. Where multiple LYN variants occurred in the same patient, all were retained at the variant level and then collapsed to a patient-level class: Truncating: any nonsense, frameshift, or canonical splice-site event present (patient labeled "truncating" irrespective of co-occurring missense variants). Missense-only: one or more missense variants with no truncating/splice events. Composite strings (e.g., multiple amino-acid changes) were parsed into constituent events; classification defaulted to truncating if any component met truncation criteria. 2.4. Domain Mapping For each variant, we extracted the codon position (when determinable) and mapped it to canonical Src-family modules: SH4/Unique (N-terminus), SH3, SH2, SH2-kinase linker, and kinase domain (activation loop noted as a subregion within the kinase domain). Splice variants without a resolvable codon number were labeled splice; frameshifts were assigned to the position of the first impacted codon. Variants lacking a reliable position were labeled unknown for domain analyses. Multi-hit policy. We prespecified that if a tumor harbored variants mapping to multiple domains, it would be collapsed to the most proximal domain. In this dataset, no tumors carried variants in more than one domain; thus, domain collapsing was not required. Unknown mapping. Variants without a resolvable protein position (e.g., splice events without a defined codon, ambiguous frameshifts) were classified as “unknown domain” and excluded from domain-level tests but retained in overall LYN-mutant analyses. 2.5. Statistical Analysis For binary comparisons, we used two-sided Fisher's exact tests and reported odds ratios (ORs) with exact 95% confidence intervals. Where any cell count was zero, we additionally reported the Haldane-Anscombe corrected OR for interpretability. Alongside pooled Fisher’s tests, we considered study-stratified Mantel–Haenszel and random-effects pooling, but did not report these due to sparse per-study cells and unstable estimates. Given small sample sizes, we fit a Firth bias-reduced logistic model with CNS(+) as the dependent variable and LYN mutation class (truncating vs missense) as the covariate. Each patient was mapped to domain presence (yes/no) for SH4/Unique, SH3, SH2, linker, and kinase domains, and per-domain Fisher's tests were conducted against CNS status. Domain-presence screens were considered hypothesis-generating; all reported p-values were Benjamini-Hochberg FDR-adjusted. Analyses were performed in R (version 4.3.2) using stats::fisher.test and logistf packages. All tests were two-sided with α = 0.05 unless stated otherwise. Effect sizes and their uncertainty were prioritized over thresholded significance. To account for between-study heterogeneity in sampling and CNS ascertainment, where estimable regression models included a study fixed effect (dummy-coded). For binary comparisons, alongside pooled Fisher’s exact tests, we considered MH and RE pooling, but did not report due to sparse per-study cells. Where appropriate, we performed random-effects inverse-variance meta-analysis of study-specific log-ORs. The prespecified Firth bias-reduced logistic model was extended to include the study effect. Two-sided α=0.05 unless stated. 2.5.1. P-Values and Confidence Intervals Two-sided p-values for 2 x 2 comparisons were obtained using conditional Fisher's exact test. Alongside pooled Fisher’s tests, we considered Mantel–Haenszel and random-effects pooling, but did not report these due to sparse per-study cells and unstable estimates. For the Firth bias-reduced logistic model (CNS(+) ~ truncating), we report profile penalized-likelihood 95% CIs for coefficients (reported as OR = e β ) and penalized likelihood-ratio test (LRT) p-values. Wald intervals were not used. 2.5.2. Sensitivity Analyses Sensitivity analyses included rare-variant thresholding (excluding singleton somatic LYN variants or those below a pre-defined frequency cut-off). For each scenario we re-fit Fisher and Firth models and report OR (95% CI) and p-values, emphasizing effect direction and precision rather than thresholded significance. 2.6. Variant-Level Clinical Annotation For each observed LYN variant, we extracted OncoKB (23, 24) (accessed 2025-09-08) annotations and integrated into the analysis workbook: oncogenicity statements (e.g., unknown oncogenic effect, likely oncogenic) and therapy notes (presence/absence of FDA-approved or compendium-listed options relevant to the alteration and disease context). 2.7. Functional enrichment We queried STRING (Homo sapiens; accessed 2025-09-11) seeding LYN with active sources: experiments, databases, co-expression; minimum required interaction score ≥ 0.70. Over-representation was tested against a genome-wide background with Benjamini–Hochberg FDR. We display top non-redundant terms with q < 0.05. 3. Results 3.1 Cohort Overview and Ascertainment of LYN Status The pooled analytic cohort comprised 5,947 unique breast cancer samples from 29 independent studies after de-duplication and quality control. LYN status was determinable in samples where panel coverage included LYN ; CNS metastasis was harmonized using a controlled vocabulary across studies (Figure 3). 3.2 Association Between LYN Mutation (Present vs WT) and CNS Metastasis In the full dataset (n = 5,947), CNS metastasis occurred in 5/46 LYN -mutated tumors (10.9%) versus 110/5,901 LYN wild-type tumors (1.9%), yielding an odds ratio = 6.42 (95% CI, 2.49–16.56) with a two-sided Fisher’s exact p = 0.0018. This suggests a statistically significant excess of CNS involvement among LYN -mutant cases (Figure 4). 3.3 Domain-Resolved Distribution of LYN Mutations by CNS Status Overall heterogeneity across domains was significant (χ² p ≈ 0.014), indicating that the location of the LYN alteration may influence CNS risk. One-versus-rest contrasts (Benjamini–Hochberg FDR control)(Figure 5): UNIQUE/SH4 (N-terminus): 3/3 in-domain cases were CNS-positive (100.0%), versus 2/16 in the remainder (12.5%). The 2×2 table contains a zero cell, yielding an infinite exact odds ratio; using a Haldane–Anscombe continuity correction for interpretability gives OR ≈ 46.2. Unadjusted p ≈ 0.010; FDR q ≈ 0.052. For zero-cell tables, we report qualitative inference from exact testing and, for interpretability, continuity-corrected odds ratios (Haldane–Anscombe). Multiple testing for domain-level one-versus-rest contrasts used BH-FDR (UNIQUE q ≈ 0.052; others not significant). SH2: 0/4 in-domain CNS-positive (0.0%) versus 5/15 in the remainder (33.3%); Fisher p ≈ 0.53. Continuity-corrected OR ≈ 0.21. SH3: 0/1 in-domain CNS-positive (0.0%) versus 5/18 in the remainder (27.8%); Fisher p ≈ 1.00. Continuity-corrected OR ≈ 0.82. Kinase domain: 2/11 in-domain CNS-positive (18.2%) versus 3/8 in the remainder (37.5%); Fisher p ≈ 0.60. OR ≈ 0.37. SH2–kinase linker: No in-domain observations; not evaluable. 3.4 Mutation Class (Truncating vs Missense) by CNS Status Among LYN -mutant cases, CNS metastasis was observed in 3/38 (7.9%) with missense variants and 2/7 (28.6%) with truncating variants. A direct Fisher’s exact comparison of these class-specific rates did not reach significance (two-sided p = 0.166) (Figure 6). Framed alternatively as the proportion of truncating variants among CNS-positive vs CNS-negative cases, the truncating share was 2/5 (40.0%) vs 6/44 (13.6%), yielding OR = 4.22 (exact 95% CI, 0.58–30.75; two-sided Fisher p = 0.182). Thus, although the point estimate exceeds 1, the confidence interval is wide and includes the null under the available sample size. In a prespecified Firth logistic model for CNS(+) ~ truncating (vs missense) restricted to LYN -mutant tumors, estimates were directionally consistent with the Fisher analysis but non-significant; profile penalized-likelihood 95% CIs were wide, reflecting sparse counts (particularly truncating, n = 7). 4. Discussion 4.1. Principal findings In this pooled, retrospectively harmonized multi-study cohort, somatic LYN mutation was associated with a higher frequency of CNS metastasis compared with LYN wild type: 10.9% (5/46) versus 1.9% (110/5,901), respectively (OR = 6.42; 95% CI, 2.49–16.56; two-sided Fisher p = 0.0018; n = 5,947). Within the LYN -mutant subset with resolvable domain mapping, domain distributions differed by CNS status (omnibus χ² p ≈ 0.014). In one-versus-rest contrasts, the SH4/Unique (N-terminus) was represented among CNS-positive cases (3/3) but not among CNS-negative entries for that domain, whereas other modules (SH2, SH3, kinase) did not show enrichment after multiplicity control (UNIQUE unadjusted p ≈ 0.010; FDR q ≈ 0.052; Haldane–Anscombe OR ≈ 46.2, acknowledging a zero cell and n = 3). By mutation class, truncating events were proportionally higher among CNS-positive than CNS-negative LYN mutants (40.0% vs 13.6%), but this comparison did not reach statistical significance (OR = 4.22; exact 95% CI, 0.58–30.75; Fisher p = 0.182). Collectively, these observations support a domain-aware pattern relating LYN variation to CNS involvement, with an N-terminal signal that is striking but sample-limited and therefore hypothesis-generating. 4.2. Biological context and plausibility The SH4/Unique segment of Src-family kinases governs membrane targeting via N-terminal lipidation and basic residue interactions with specific microdomains. An enrichment signal at this module—albeit based on very small counts—is biologically coherent with steps requisite for CNS spread: endothelial adhesion, BBB transmigration, and early perivascular survival, processes that depend on precise spatial coupling of integrins/FAK, RTKs, and cytoskeletal machinery. By contrast, the kinase domain was represented across CNS strata without clear differential enrichment, suggesting that localization/context—rather than catalytic-core involvement alone—may distinguish CNS behavior for LYN -mutant tumors in this dataset. These data motivate functional tests of how N-terminal perturbations rewire receptor–adhesion complexes and barrier interactions in brain endothelium models. 4.3. Interpreting the mutation-class signal The truncating-versus-missense comparison yielded point estimates > 1 but non-significant p values under exact testing (p = 0.166 for class-specific rates; p = 0.182 for the truncating share analysis), with wide CIs reflecting sparse truncating counts (n = 7) and only five CNS-positive LYN -mutant cases overall. These results are compatible with either a true but imprecise effect or no effect; as such, they should not be over-interpreted absent replication or larger pooled analyses and, ideally, orthogonal functional evidence. 4.4. Mechanistic considerations: LYN , CNS tropism, and the tumor microenvironment LYN sits at a nodal position connecting receptor tyrosine kinases, integrins, immune receptors and cytoskeletal scaffolds. Perturbations of this hub—whether by mutation, truncation, or domain-specific rewiring—could plausibly shift adhesion, motility and survival programs in ways that favor CNS dissemination. Beyond tumor-intrinsic effects, LYN also operates in endothelial and immune compartments, suggesting multi-cellular routes to brain tropism ( 5 , 11 , 12 , 14 , 21 ). A first mechanistic axis concerns membrane targeting and nanoscale localization. The SH4/Unique N-terminus controls myristoylation/palmitoylation-dependent membrane association and partitioning into lipid microdomains; even small changes here can redirect the kinase to distinct receptor/scaffold neighborhoods. In principle, N-terminal alterations could bias LYN toward complexes that enhance endothelial adhesion and barrier engagement, or impair autoinhibitory clamping, thereby lowering the activation threshold at the cell periphery ( 5 , 16 , 17 , 21 ). Such relocalization would be expected to remodel proximal phosphorylation of focal-adhesion adaptors and junctional proteins. Second, adhesion–cytoskeleton coupling provides a convergent route to motility and diapedesis. LYN phosphorylates focal-adhesion components (e.g., BCAR1/CAS, NEDD9/HEF1) and interfaces with integrin/FAK signaling to accelerate adhesion turnover and directional migration ( 12 ). In parallel, cross-talk with ERBB/EGFR and downstream PI3K–AKT and MAPK cascades can increase survival of intravasating or circulating cells, potentially enlarging the pool of cells that can reach and interrogate the brain endothelium ( 12 , 13 ). Third, endothelial and barrier biology may link LYN to the blood–brain barrier (BBB). LYN activity has been implicated in regulating endothelial activation, leukocyte adhesion and transendothelial migration, processes that mirror the steps required for tumor cell extravasation ( 5 , 21 ). If LYN -dependent signaling modulates endothelial permeability or the presentation of adhesion molecules under inflammatory cues, tumor cells with LYN pathway perturbations could gain context-dependent advantages at the BBB interface ( 14 , 21 ). Fourth, immune regulation offers a microenvironmental bridge. LYN can act as a negative regulator by phosphorylating ITIM-containing receptors and recruiting phosphatases (SHP-1/SHP-2/SHIP-1), damping immune activation in B cells and myeloid lineages while still permitting context-specific activation ( 11 ). In a tumor setting, such tuning might blunt anti-tumor immunity locally or systemically, alter neutrophil/monocyte trafficking, and reshape cytokine milieus that influence BBB function and niche formation ( 14 ). These effects need not be restricted to tumor cells; host leukocytes with LYN -dependent signaling could indirectly facilitate metastatic seeding. Fifth, chemokine and growth-factor axes downstream of LYN (e.g., CXCR4, EPOR, KIT) point to additional, testable circuits. CXCL12–CXCR4 gradients are operative at vascular and neural interfaces; LYN -mediated propagation of such signals could coordinate chemotaxis with adhesion mechanics. Likewise, LYN ’s links to platelet and erythroid signaling raise the possibility that platelet cloaking or altered rheology contributes to intravascular survival and endothelial docking, although these ideas remain speculative without matched functional data. Together, these lines of reasoning generate concrete predictions. If N-terminal/domain context matters for CNS tropism, LYN -mutant tumors that seed the CNS should show (i) altered subcellular localization and membrane microdomain occupancy; (ii) rewired phosphoproteomic signatures at focal adhesions and junctions; (iii) enhanced adhesion/diapedesis across human brain microvascular endothelial monolayers under shear; and (iv) microenvironmental differences in endothelial activation and immune infiltration. These can be evaluated in isogenic systems expressing representative SH4/Unique versus kinase-core variants, using live-cell imaging, microfluidic BBB models, and unbiased phospho-proteomics; in clinical material, spatial IHC or transcriptomic deconvolution (B-cell and myeloid signatures, adhesion/chemokine programs) could probe whether the LYN –CNS signal co-travels with microenvironmental features. Finally, the therapeutic implications are purposely cautious. Pan-SFK inhibition has produced mixed results and may be limited by BBB penetration and pathway redundancy; conversely, domain-driven rewiring suggests combinations—e.g., SRC-family inhibitors with integrin/FAK blockade or anti-inflammatory modulators—could be more effective in biologically defined subsets ( 12 , 19 , 21 ). At present, our findings are hypothesis-generating; establishing independence from canonical clinical drivers and embedding the LYN signal within a microenvironmental framework will require subtype-balanced, time-resolved datasets and mechanistic validation across the tumor–endothelium–immune triad. 4.4. Methodological strengths This work adheres to several practices that increase inferential robustness for rare alterations and sparse endpoints: Prespecified endpoint harmonization using a controlled CNS vocabulary with no imputation for missing site fields; Exact procedures appropriate to small cells (two-sided Fisher tests, exact/mid-P intervals where feasible) and BH-FDR control for domain screens; A priori domain mapping and patient-level collapsing rules; Multi-study pooling with de-duplication across patient/sample identifiers to reduce double counting. 4.5. Limitations Important caveats temper interpretation. The retrospective design risks selection and information bias from clinically sequenced cohorts and free-text metastasis fields (despite normalization). The LYN -mutant denominator is small (n = 46; CNS-positive n = 5), yielding imprecise estimates with wide CIs and limiting multivariable adjustment. Key clinical/genomic covariates (subtype, treatment, stage, co-mutations, copy number/expression, TMB) were incompletely available, leaving residual confounding or effect modification. CNS event timing (baseline vs follow-up) was inconsistently captured, precluding temporal inference. Finally, heterogeneity in panel coverage and site capture across studies may introduce between-study variability only partially addressed by pooled exact analyses. 4.6. Implications and next steps These findings elevate domain context as a key lens for assessing LYN variation and CNS risk. Next steps: External validation: Well-annotated cohorts with harmonized CNS fields and sufficient LYN -mut/CNS + cases to refine estimates and enable covariate-adjusted models (e.g., subtype-adjusted Firth logistic; study-stratified or random-effects meta-analysis). Temporal curation: Time-stamp CNS events to separate baseline from incident metastases and test lead-lag relationships with LYN status. Mechanistic studies: Isogenic models comparing SH4/Unique vs kinase-core/regulatory-head perturbations, probing membrane targeting, endothelial adhesion, and diapedesis in reductionist and microphysiological BBB systems. Signaling/interactome profiling: Determine whether N-terminal alterations reprogram RTK-integrin coupling or scaffold composition to favor CNS tropism. Clinical pipelines: Domain-aware curation to standardize future aggregation/meta-analysis and flag putatively high-risk patterns for prospective surveillance. 5. Conclusion In a harmonized, multi-study breast cancer cohort, any somatic LYN mutation was associated with increased odds of CNS metastasis (10.9% vs 1.9%; OR = 6.42; 95% CI, 2.49–16.56; p = 0.0018). Among LYN -mutant tumors, domain distributions differed by CNS status, with a small-sample enrichment signal at the SH4/Unique N-terminus (FDR-borderline), while other modules did not show significant enrichment. The truncating-versus-missense comparison was directionally positive but statistically non-significant under sparse counts, and should be viewed as exploratory. Overall, the data elevate domain context—particularly N-terminal membrane-targeting logic—as a mechanistically plausible axis for future validation and functional interrogation. Given the limited power and heterogeneous annotations, immediate therapeutic implications are premature, especially considering the challenges of CNS drug delivery and the mixed performance of unselected SFK inhibition. Prospective validation in larger, uniformly annotated datasets, coupled with domain-informed experiments centered on membrane localization and BBB traversal, will be critical to determine whether LYN domain patterns can inform risk modeling or biomarker development in CNS involvement of breast cancer. Findings are hypothesis-generating and require independent validation in larger, uniformly annotated cohorts before any clinical application. Abbreviations CNS, BBB, SFK, RTK, EMT, OR, CI, FDR, IRB, NST, SNV. Declarations Ethics approval and consent to participate: This study analyzed only publicly available, de-identified data aggregated from multiple studies and portals. There was no interaction with human participants and no access to identifiable private information. In accordance with applicable regulations and institutional policies, this secondary analysis of de-identified public data is not considered human subjects research; institutional review board oversight and informed consent were not required. Reporting follows STROBE. Consent for publication: Not applicable. Availability of data and materials: All datasets analyzed in this study are publicly accessible from cBioPortal for Cancer Genomics. Direct study links and identifiers are provided below. We also used OncoKB for variant-level annotations and STRING for enrichment analyses; access dates are indicated. cBioPortal for Cancer Genomics — breast_msk_2025: https://www.cbioportal.org/study/summary?id=breast_msk_2025 cBioPortal for Cancer Genomics — brca_tcga: https://www.cbioportal.org/study/summary?id=brca_tcga cBioPortal for Cancer Genomics — breast_ink4_msk_2021: https://www.cbioportal.org/study/summary?id=breast_ink4_msk_2021 cBioPortal for Cancer Genomics — breast_msk_2018: https://www.cbioportal.org/study/summary?id=breast_msk_2018 cBioPortal for Cancer Genomics — brca_msk_2025: https://www.cbioportal.org/study/summary?id=brca_msk_2025 cBioPortal for Cancer Genomics — brca_mbcproject_2022: https://www.cbioportal.org/study/summary?id=brca_mbcproject_2022 cBioPortal for Cancer Genomics — brca_igr_2015: https://www.cbioportal.org/study/summary?id=brca_igr_2015 cBioPortal for Cancer Genomics — brca_smc_2018: https://www.cbioportal.org/study/summary?id=brca_smc_2018 cBioPortal for Cancer Genomics — brca_aurora_2023: https://www.cbioportal.org/study/summary?id=brca_aurora_2023 cBioPortal for Cancer Genomics — breast_cptac_gdc: https://www.cbioportal.org/study/summary?id=breast_cptac_gdc cBioPortal for Cancer Genomics — brca_cptac_2020: https://www.cbioportal.org/study/summary?id=brca_cptac_2020 cBioPortal for Cancer Genomics — brca_bccrc_xenograft_2014: https://www.cbioportal.org/study/summary?id=brca_bccrc_xenograft_2014 cBioPortal for Cancer Genomics — brca_broad: https://www.cbioportal.org/study/summary?id=brca_broad cBioPortal for Cancer Genomics — brca_sanger: https://www.cbioportal.org/study/summary?id=brca_sanger cBioPortal for Cancer Genomics — brca_fuscc_2020 (n=69): https://www.cbioportal.org/study/summary?id=brca_fuscc_2020 cBioPortal for Cancer Genomics — brca_bccrc: https://www.cbioportal.org/study/summary?id=brca_bccrc cBioPortal for Cancer Genomics — brca_mapk_hp_msk_2021: https://www.cbioportal.org/study/summary?id=brca_mapk_hp_msk_2021 cBioPortal for Cancer Genomics — brca_mbcproject_wagle_2017: https://www.cbioportal.org/study/summary?id=brca_mbcproject_wagle_2017 cBioPortal for Cancer Genomics — brca_pareja_msk_2020: https://www.cbioportal.org/study/summary?id=brca_pareja_msk_2020 cBioPortal for Cancer Genomics — brca_dfci_2020: https://www.cbioportal.org/study/summary?id=brca_dfci_2020 cBioPortal for Cancer Genomics — bfn_duke_nus_2015: https://www.cbioportal.org/study/summary?id=bfn_duke_nus_2015 cBioPortal for Cancer Genomics — mbc_msk_2021 https://www.cbioportal.org/study/summary?id=mbc_msk_2021 cBioPortal for Cancer Genomics — ilc_msk_2023: https://www.cbioportal.org/study/summary?id=ilc_msk_2023 cBioPortal for Cancer Genomics — acbc_mskcc_2015: https://www.cbioportal.org/study/summary?id=acbc_mskcc_2015 cBioPortal for Cancer Genomics — breast_alpelisib_2020: https://www.cbioportal.org/study/summary?id=breast_alpelisib_2020 cBioPortal for Cancer Genomics — brca_jup_msk_2020: https://www.cbioportal.org/study/summary?id=brca_jup_msk_2020 cBioPortal for Cancer Genomics — brca_hta9_htan_2022: https://www.cbioportal.org/study/summary?id=brca_hta9_htan_2022 cBioPortal for Cancer Genomics — brca_tcga_gdc: https://www.cbioportal.org/study/summary?id=brca_tcga_gdc cBioPortal for Cancer Genomics — brca_tcga_pan_can_atlas_2018: https://www.cbioportal.org/study/summary?id=brca_tcga_pan_can_atlas_2018 Other resources: OncoKB — https://www.oncokb.org/ (accessed 2025-09-08) STRING — https://string-db.org/ (accessed 2025-09-11) Competing interests: The authors declare no competing interests. Funding: No specific funding was received for this work. Authors’ contributions: E.K.Ç.: Conceptualization; Study design/Methodology; Data curation; Resources; Supervision; Writing—original draft. B.Ç.: Software; Formal analysis/Statistics; Visualization; Writing—original draft. All authors read and approved the final manuscript. Acknowledgements: We gratefully acknowledge the cBioPortal for Cancer Genomics team for providing an open platform to access and analyze cancer genomics data, and the STRING consortium for protein association networks and enrichment tools used in this study. We also thank the contributing investigators, institutions, and—most importantly—the patients—whose data made this research possible. Interpretations and conclusions are solely those of the authors and do not necessarily reflect the views of the data contributors. References Raghavendra AS, Ibrahim NK. Breast Cancer Brain Metastasis: A Comprehensive Review. JCO Oncol Pract. 2024;20(10):1348-59. Stavrou E, Winer EP, Lin NU. How we treat HER2-positive brain metastases. ESMO Open. 2021;6(5):100256. Boire A, Brastianos PK, Garzia L, Valiente M. Brain metastasis. Nat Rev Cancer. 2020;20(1):4-11. Soung YH, Clifford JL, Chung J. Crosstalk between integrin and receptor tyrosine kinase signaling in breast carcinoma progression. BMB Rep. 2010;43(5):311-8. Han J, Zhang G, Welch EJ, Liang Y, Fu J, Vogel SM, et al. A critical role for Lyn kinase in strengthening endothelial integrity and barrier function. Blood. 2013;122(25):4140-9. Tornillo G, Knowlson C, Kendrick H, Cooke J, Mirza H, Aurrekoetxea-Rodríguez I, et al. Dual Mechanisms of LYN Kinase Dysregulation Drive Aggressive Behavior in Breast Cancer Cells. Cell Rep. 2018;25(13):3674-92.e10. Parsons SJ, Parsons JT. Src family kinases, key regulators of signal transduction. Oncogene. 2004;23(48):7906-9. Szklarczyk D, Kirsch R, Koutrouli M, Nastou K, Mehryary F, Hachilif R, et al. The STRING database in 2023: protein-protein association networks and functional enrichment analyses for any sequenced genome of interest. Nucleic Acids Res. 2023;51(D1):D638-d46. Choi YL, Bocanegra M, Kwon MJ, Shin YK, Nam SJ, Yang JH, et al. LYN is a mediator of epithelial-mesenchymal transition and a target of dasatinib in breast cancer. Cancer Res. 2010;70(6):2296-306. Safran M, Rosen N, Twik M, BarShir R, Stein TI, Dahary D, et al. The GeneCards Suite. In: Abugessaisa I, Kasukawa T, editors. Practical Guide to Life Science Databases. Singapore: Springer Nature Singapore; 2021. p. 27-56. Doron H, Pukrop T, Erez N. A Blazing Landscape: Neuroinflammation Shapes Brain Metastasis. Cancer Research. 2019;79(3):423-36. Wheeler SE, Morariu EM, Bednash JS, Otte CG, Seethala RR, Chiosea SI, et al. Lyn kinase mediates cell motility and tumor growth in EGFRvIII-expressing head and neck cancer. Clin Cancer Res. 2012;18(10):2850-60. Okada M. Regulation of the SRC family kinases by Csk. Int J Biol Sci. 2012;8(10):1385-97. Sicheri F, Moarefi I, Kuriyan J. Crystal structure of the Src family tyrosine kinase Hck. Nature. 1997;385(6617):602-9. Boggon TJ, Eck MJ. Structure and regulation of Src family kinases. Oncogene. 2004;23(48):7918-27. Koboldt DC, Fulton RS, McLellan MD, Schmidt H, Kalicki-Veizer J, McMichael JF, et al. Comprehensive molecular portraits of human breast tumours. Nature. 2012;490(7418):61-70. Zhang S, Yu D. Targeting Src family kinases in anti-cancer therapies: turning promise into triumph. Trends in Pharmacological Sciences. 2012;33(3):122-8. Wang Y, Ye F, Liang Y, Yang Q. Breast cancer brain metastasis: insight into molecular mechanisms and therapeutic strategies. British Journal of Cancer. 2021;125(8):1056-67. Arvanitis CD, Ferraro GB, Jain RK. The blood-brain barrier and blood-tumour barrier in brain tumours and metastases. Nat Rev Cancer. 2020;20(1):26-41. de Bruijn I, Kundra R, Mastrogiacomo B, Tran TN, Sikina L, Mazor T, et al. Analysis and Visualization of Longitudinal Genomic and Clinical Data from the AACR Project GENIE Biopharma Collaborative in cBioPortal. Cancer Res. 2023;83(23):3861-7. Gao J, Aksoy BA, Dogrusoz U, Dresdner G, Gross B, Sumer SO, et al. Integrative analysis of complex cancer genomics and clinical profiles using the cBioPortal. Sci Signal. 2013;6(269):pl1. Cerami E, Gao J, Dogrusoz U, Gross BE, Sumer SO, Aksoy BA, et al. The cBio cancer genomics portal: an open platform for exploring multidimensional cancer genomics data. Cancer Discov. 2012;2(5):401-4. Suehnholz SP, Nissan MH, Zhang H, Kundra R, Nandakumar S, Lu C, et al. Quantifying the Expanding Landscape of Clinical Actionability for Patients with Cancer. Cancer Discovery. 2024;14(1):49-65. Chakravarty D, Gao J, Phillips S, Kundra R, Zhang H, Wang J, et al. OncoKB: A Precision Oncology Knowledge Base. JCO Precision Oncology. 2017(1):1-16. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 05 Dec, 2025 Read the published version in BMC Cancer → Version 1 posted Editorial decision: Revision requested 31 Oct, 2025 Reviews received at journal 21 Oct, 2025 Reviews received at journal 17 Oct, 2025 Reviewers agreed at journal 10 Oct, 2025 Reviewers agreed at journal 08 Oct, 2025 Reviews received at journal 08 Oct, 2025 Reviewers agreed at journal 07 Oct, 2025 Reviewers invited by journal 07 Oct, 2025 Editor assigned by journal 07 Oct, 2025 Editor invited by journal 18 Sep, 2025 Submission checks completed at journal 17 Sep, 2025 First submitted to journal 17 Sep, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7597855","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":531590486,"identity":"e15ca3a0-7b69-4121-a0c2-48ce5d2d2a33","order_by":0,"name":"Elif Kardelen Çağdaş","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABM0lEQVRIie2QMWuDQBiG7zg4lyuuF6z9DRZBCh36V3Sxi05dbfAQdEl11aX9C2bJWuHodQlkdSqGQIZODQHJJNUUOgSFjIX6DO/dwffw3h0AIyN/EQn6Ox8A6Es/Z+gDVLQrvRxUEGTZUUG/CjY7hQwrAJ4qROs2g4qMEGOp96EmCInq4D2qSbLcf5beDQESf8t7lEkAGcvFg54F+P56Jt71tHQXt45oL0Zsu+xRNA6jdYVNK+fEUAAWll9eLHQHtwolRp9yx9uWqjGtVy7XCmiE9bJabnWnGVa09sfYPGxbEMEKDD0rLxy0ccNhhXYtWWzqKcfG5Cku9HlpG8iNKcEDb5GjaM1mtakmUbClh3qqPq/4Zu/U0ytZ4qJPOYV3gekxzxjvmHaBvs6cHhkZGfkffAOxKmwIp3EfWgAAAABJRU5ErkJggg==","orcid":"","institution":"Ankara University","correspondingAuthor":true,"prefix":"","firstName":"Elif","middleName":"Kardelen","lastName":"Çağdaş","suffix":""},{"id":531590487,"identity":"92aae0eb-d7f4-459f-b246-2184a1e37d6e","order_by":1,"name":"Berkay Çağdaş","email":"","orcid":"","institution":"Afyonkarahisar State Hospital","correspondingAuthor":false,"prefix":"","firstName":"Berkay","middleName":"","lastName":"Çağdaş","suffix":""}],"badges":[],"createdAt":"2025-09-12 07:53:18","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7597855/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7597855/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1186/s12885-025-15398-w","type":"published","date":"2025-12-05T15:57:42+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":93920889,"identity":"36366131-2527-48b7-b2de-0fc598cdf021","added_by":"auto","created_at":"2025-10-20 09:42:37","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":126499,"visible":true,"origin":"","legend":"","description":"","filename":"Manuscript2.docx","url":"https://assets-eu.researchsquare.com/files/rs-7597855/v1/f4ffa97e383d25a2781de489.docx"},{"id":93920895,"identity":"a8b4bd04-098a-41f6-920a-b435437c7793","added_by":"auto","created_at":"2025-10-20 09:42:37","extension":"json","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":10290,"visible":true,"origin":"","legend":"","description":"","filename":"e373743578c640ef9e77f1c836763638.json","url":"https://assets-eu.researchsquare.com/files/rs-7597855/v1/6882a7ea86df809e5d525d1e.json"},{"id":93921864,"identity":"155d14ed-892c-4ad0-b01a-690c8cb533b0","added_by":"auto","created_at":"2025-10-20 09:50:38","extension":"xml","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":105015,"visible":true,"origin":"","legend":"","description":"","filename":"e373743578c640ef9e77f1c8367636381enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7597855/v1/7de8b90d770214c1c9a1b6ac.xml"},{"id":93921862,"identity":"a154a7e0-40ec-4d0f-8d01-14ad59401efc","added_by":"auto","created_at":"2025-10-20 09:50:37","extension":"jpg","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":27343,"visible":true,"origin":"","legend":"","description":"","filename":"Figure1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7597855/v1/370ced366645c092d55772cc.jpg"},{"id":93923043,"identity":"545c3d16-8558-460a-8404-beb87bdf6859","added_by":"auto","created_at":"2025-10-20 09:58:37","extension":"jpg","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":673532,"visible":true,"origin":"","legend":"","description":"","filename":"Figure2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7597855/v1/2703efb8f62cd80ae1e6811c.jpg"},{"id":93920902,"identity":"109a2c20-172f-499a-bc16-69fa334a5cda","added_by":"auto","created_at":"2025-10-20 09:42:37","extension":"jpg","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":123805,"visible":true,"origin":"","legend":"","description":"","filename":"Figure3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7597855/v1/fa935cabf0e899ff90686a3f.jpg"},{"id":93923499,"identity":"2cdeca9d-fcb0-4f2d-938f-d16d789eb937","added_by":"auto","created_at":"2025-10-20 10:06:37","extension":"jpg","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":253097,"visible":true,"origin":"","legend":"","description":"","filename":"Figure4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7597855/v1/d081d77b3f1edd42fa66a811.jpg"},{"id":93920899,"identity":"470a119f-7310-4126-8854-8a9fbee952fe","added_by":"auto","created_at":"2025-10-20 09:42:37","extension":"jpg","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":300053,"visible":true,"origin":"","legend":"","description":"","filename":"Figure5.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7597855/v1/5c50fc80716ffe67b9164ea8.jpg"},{"id":93923498,"identity":"f35be335-17c3-4d73-af16-bae9276a07ee","added_by":"auto","created_at":"2025-10-20 10:06:37","extension":"jpg","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":260134,"visible":true,"origin":"","legend":"","description":"","filename":"Figure6.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7597855/v1/55cbb8499a08cb075c132a38.jpg"},{"id":93920892,"identity":"c86181d3-8d53-4e12-bfe1-e70914eb28ce","added_by":"auto","created_at":"2025-10-20 09:42:37","extension":"png","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":9324,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure1.png","url":"https://assets-eu.researchsquare.com/files/rs-7597855/v1/e6b0453f2a5c82f97070570c.png"},{"id":93920905,"identity":"3428b349-e00d-4ef1-aec7-7aed17c23cb3","added_by":"auto","created_at":"2025-10-20 09:42:38","extension":"png","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":139308,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure2.png","url":"https://assets-eu.researchsquare.com/files/rs-7597855/v1/33b61eaae51624aaf4c762ef.png"},{"id":93923041,"identity":"6acfe58b-af66-48a6-aaa1-1a5cb4838536","added_by":"auto","created_at":"2025-10-20 09:58:37","extension":"png","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":34961,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure3.png","url":"https://assets-eu.researchsquare.com/files/rs-7597855/v1/0399c911b765248d8d993b59.png"},{"id":93921857,"identity":"dabb9dd6-b623-45e2-ba68-a98095556cc9","added_by":"auto","created_at":"2025-10-20 09:50:37","extension":"png","order_by":18,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":38320,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure4.png","url":"https://assets-eu.researchsquare.com/files/rs-7597855/v1/e6f4b9a3bf21082e19cd455b.png"},{"id":93920903,"identity":"32b74c3a-2805-43a0-b4d7-1a8923bbce03","added_by":"auto","created_at":"2025-10-20 09:42:38","extension":"png","order_by":19,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":45502,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure5.png","url":"https://assets-eu.researchsquare.com/files/rs-7597855/v1/3f3f4282064835964b23b61d.png"},{"id":93920907,"identity":"286aa03e-97d5-4c23-a401-c469259840f3","added_by":"auto","created_at":"2025-10-20 09:42:38","extension":"png","order_by":20,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":35616,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure6.png","url":"https://assets-eu.researchsquare.com/files/rs-7597855/v1/af71a34687186bb4b0864ef7.png"},{"id":93920910,"identity":"4e40db37-2bd1-44b5-9d55-e4d8de653071","added_by":"auto","created_at":"2025-10-20 09:42:38","extension":"xml","order_by":21,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":106486,"visible":true,"origin":"","legend":"","description":"","filename":"e373743578c640ef9e77f1c8367636381structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7597855/v1/d5253cfe8fae9dec61150731.xml"},{"id":93921865,"identity":"49957e2e-e5db-4b42-9400-9570667cd2b2","added_by":"auto","created_at":"2025-10-20 09:50:38","extension":"html","order_by":22,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":114907,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7597855/v1/687e41bfdc25f83f88379bf5.html"},{"id":93921856,"identity":"e0d8315f-ec09-400f-97f9-05db1752bfe1","added_by":"auto","created_at":"2025-10-20 09:50:37","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":27343,"visible":true,"origin":"","legend":"\u003cp\u003eGO-BP enrichment of the LYN-centered STRING network (8).\u003c/p\u003e\n\u003cp\u003eTop enriched terms cluster in immune receptor signaling (B-cell/Fc/TLR), integrin–cytoskeleton/adhesion, leukocyte migration, and PI3K–AKT/MAPK cascades—consistent with LYN-linked adhesion/motility and barrier interactions relevant to CNS tropism.\u003c/p\u003e","description":"","filename":"Figure1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7597855/v1/50cd36576ab854d38085ff6b.jpg"},{"id":93920891,"identity":"0b9f75be-2d39-471b-b9e1-9540812a0040","added_by":"auto","created_at":"2025-10-20 09:42:37","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":673532,"visible":true,"origin":"","legend":"\u003cp\u003eStructural architecture of human \u003cem\u003eLYN\u003c/em\u003ekinase showing α-helices and β-sheets (8, 9).\u003c/p\u003e\n\u003cp\u003eCartoon of full-length \u003cem\u003eLYN\u003c/em\u003e(UniProt P07948) with Src-family modules: SH4/Unique (membrane targeting), SH3, SH2, SH2–kinase linker, and the bilobed kinase domain. α-Helices are shown as coils/cylinders, β-sheets as arrows; the N-terminal SH4/Unique is largely disordered in many models. The regulatory head (SH3/SH2) mediates autoinhibition via the proline-rich linker and the C-terminal inhibitory tyrosine, whereas the kinase activation loop tyrosine marks the active state. pLDDT coloring (Model Confidence: Very high (pLDDT \u0026gt; 90), Confident (90 \u0026gt; pLDDT \u0026gt; 70), Low (70 \u0026gt; pLDDT \u0026gt; 50), Very low (pLDDT \u0026lt; 50)) indicates model confidence. This view serves as a scaffold to map patient variants and interpret domain-specific signals in the CNS-metastasis analyses.\u003c/p\u003e","description":"","filename":"Figure2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7597855/v1/9a9d4413aef5e8c6a946c52c.jpg"},{"id":93920887,"identity":"7a82a3f9-6365-4764-bebd-408e9c49011d","added_by":"auto","created_at":"2025-10-20 09:42:37","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":123805,"visible":true,"origin":"","legend":"\u003cp\u003eCohort assembly and analytic subsets.\u003c/p\u003e\n\u003cp\u003eFrom 17,605 breast cancer samples in cBioPortal (accessed September 2025), we retained 16,881 with somatic mutation profiling; restricted to invasive carcinoma of no special type (NST) (n = 14,050); and, after de-duplication across studies and requiring \u003cem\u003eLYN\u003c/em\u003ecoverage, included 5,947 non-overlapping cases in the pooled analysis. Of these, 46 harbored somatic \u003cem\u003eLYN\u003c/em\u003e mutations and 5,901 were LYN wild type. CNS metastasis—captured as ever vs never (event dates unavailable)—was present in 5/46 \u003cem\u003eLYN\u003c/em\u003e-mutant and 110/5,901 \u003cem\u003eLYN\u003c/em\u003e–wild-type tumors. Exclusions by step: lack of somatic profiling (n = 724), non-NST histology or missing histology (n = 2,831), and duplicate patients or missing \u003cem\u003eLYN\u003c/em\u003e sequencing (n = 8,103).\u003c/p\u003e","description":"","filename":"Figure3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7597855/v1/e05fd74b54c40548d2811ddd.jpg"},{"id":93921859,"identity":"af06a2ea-041b-4b79-8ff3-a57b11b1836b","added_by":"auto","created_at":"2025-10-20 09:50:37","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":253097,"visible":true,"origin":"","legend":"\u003cp\u003eCNS metastasis frequency by LYN mutation status.\u003c/p\u003e\n\u003cp\u003eBar plot showing the proportion of tumors with CNS metastasis in the pooled NST cohort (n = 5,947). Among \u003cem\u003eLYN\u003c/em\u003e-mutant cases, 5/46 (10.9%) had CNS metastasis; among \u003cem\u003eLYN\u003c/em\u003e wild type cases, 110/5,901 (1.9%) had CNS metastasis. The difference corresponds to an odds ratio = 6.42 (95% CI, 2.49–16.56; two-sided Fisher’s exact p = 0.0018). Endpoint recorded as ever vs never CNS involvement; event dates unavailable (no time-to-event).\u003c/p\u003e","description":"","filename":"Figure4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7597855/v1/6b0a79df634ca5e0ad3ed851.jpg"},{"id":93920893,"identity":"c0c1d0b2-7af1-42fa-8f7e-f0b44e812dec","added_by":"auto","created_at":"2025-10-20 09:42:37","extension":"jpg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":300053,"visible":true,"origin":"","legend":"\u003cp\u003eCNS metastasis by \u003cem\u003eLYN\u003c/em\u003e mutation domain.\u003c/p\u003e\n\u003cp\u003eWithin \u003cem\u003eLYN\u003c/em\u003e-mutant tumors (n = 46), bars show % CNS-positive by domain (SH4/Unique, SH3, SH2, SH2–kinase linker, kinase). Endpoint recorded as ever vs never CNS involvement; event dates unavailable (no time-to-event).\u003c/p\u003e","description":"","filename":"Figure5.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7597855/v1/5750c70a4f3de531bc2b4c96.jpg"},{"id":93920890,"identity":"0c164f4a-2fb6-40f2-b3bc-fe0ffe1424db","added_by":"auto","created_at":"2025-10-20 09:42:37","extension":"jpg","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":260134,"visible":true,"origin":"","legend":"\u003cp\u003eCNS metastasis by LYN mutation class.\u003c/p\u003e\n\u003cp\u003eWithin \u003cem\u003eLYN\u003c/em\u003e-mutant tumors (n = 46), bars show % CNS-positive for truncating vs missense variants (counts atop bars; error bars = exact 95% CIs). CNS+: 28.6% (truncating) vs 7.9% (missense); Fisher p = 0.166; OR 4.22 (95% CI 0.58–30.75). Firth logistic was directionally concordant but imprecise.\u003c/p\u003e","description":"","filename":"Figure6.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7597855/v1/83f611c154829d613895bd9a.jpg"},{"id":97724717,"identity":"2a34cc48-1ca5-4bfd-8cf1-d05eedb06f22","added_by":"auto","created_at":"2025-12-08 16:13:25","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2712224,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7597855/v1/63590372-6b37-4c43-853c-ccc950078b42.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"LYN Mutations in Breast Cancer: Association with Central Nervous System Metastasis and Domain-Level Insights","fulltext":[{"header":"Key messages","content":"\u003cp\u003e\u003cstrong\u003eWhat is already known:\u0026nbsp;\u003c/strong\u003eCNS metastasis is a major clinical problem in breast cancer; molecular determinants of brain tropism remain incompletely defined.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eWhat this study adds:\u0026nbsp;\u003c/strong\u003eAcross pooled cohorts, any \u003cem\u003eLYN\u003c/em\u003e mutation is associated with higher odds of CNS metastasis; domain context (notably SH4/Unique) may matter, albeit with small counts and FDR-borderline evidence.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eHow this might affect research/practice/policy:\u0026nbsp;\u003c/strong\u003eEncourages domain-aware \u003cem\u003eLYN\u003c/em\u003e annotation and targeted mechanistic work on membrane targeting/BBB traversal; not ready for risk stratification or treatment selection.\u003c/p\u003e"},{"header":"1. Introduction","content":"\u003cp\u003eBreast cancer remains a major cause of cancer morbidity and mortality, with a substantial share driven by CNS involvement (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e). As systemic therapies extend survival, brain and leptomeningeal metastases have greater clinical impact, yet the molecular drivers of CNS tropism are still poorly defined (\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e). Brain colonization demands tumor-intrinsic programs\u0026mdash;motility, invasion, survival\u0026mdash;plus BBB transmigration and adaptation to the neural microenvironment (\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e). Signaling modules that couple cell adhesion to RTK inputs and cytoskeletal regulation are thus strong candidates (\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e). Src family kinases\u0026mdash;especially \u003cem\u003eLYN\u003c/em\u003e\u0026mdash;sit at this nexus, integrating integrin/FAK signaling, ERBB/EGFR cross-talk, and pathways controlling EMT, cell\u0026ndash;cell junctions, and actin remodeling (\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e). Although \u003cem\u003eLYN\u003c/em\u003e copy-number/expression changes recur across cancers, the clinical relevance of specific \u003cem\u003eLYN\u003c/em\u003e variants in breast cancer, particularly for CNS involvement, remains undefined (\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e), motivating a focused analysis of \u003cem\u003eLYN\u003c/em\u003e\u0026rsquo;s structure\u0026ndash;function context and an exploratory test of whether mutation class and domain location associate with CNS tropism.\u003c/p\u003e\u003cdiv id=\"Sec2\" class=\"Section2\"\u003e\u003ch2\u003e1.1. \u003cem\u003eLYN\u003c/em\u003e Within the Src Family: Overview and Relevance to Breast Cancer\u003c/h2\u003e\u003cp\u003e\u003cem\u003eLYN\u003c/em\u003e is a non-receptor tyrosine kinase in the Src family (SFKs), conserved signal integrators that relay inputs from cell-surface receptors to control proliferation, survival, adhesion, motility, and cytoskeletal dynamics (\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e). SFKs are rapidly engaged downstream of RTKs, integrins, GPCRs, immune and cytokine receptors, and phosphorylate substrates that scaffold focal adhesions, remodel actin, and amplify mitogenic/pro-survival cascades (\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e). Because of this nodal position, small shifts in SFK activity or localization can produce outsized cellular effects. In immunity, \u003cem\u003eLYN\u003c/em\u003e frequently acts as a negative regulator yet can also activate signaling, being required for both initiation and termination of B-cell responses (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003ePhysiologically, \u003cem\u003eLYN\u003c/em\u003e tunes B-cell receptor signaling, Fc pathways in myeloid/mast cells, and tonic thresholds for activation/tolerance, acting positively or negatively depending on the complex and adaptors (\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e). Upstream inputs include BCR (CD79A/B), CD5, CD19, CD22, Fc receptors, TLR2/4, and receptors such as EPOR, KIT, MPL, CXCR4, and IL-3/IL-5/CSF2 (\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e). Outside hematopoiesis, LYN in epithelial tissues\u0026mdash;including breast cancer\u0026mdash;regulates cell\u0026ndash;matrix adhesion, migration, EMT, and, via integrin/FAK and RTK (ERBB/EGFR) coupling to PI3K\u0026ndash;AKT and RAS\u0026ndash;RAF\u0026ndash;MEK\u0026ndash;ERK, enhances survival and proliferation (\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e). It also dampens signaling by phosphorylating ITIMs to recruit SHP-1/2 and SHIP-1.\u003c/p\u003e\u003cp\u003e\u003cem\u003eLYN\u003c/em\u003e\u0026rsquo;s qualitative output depends on conformation, localization, and scaffold/phosphatase balance, explaining context-dependent effects (\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e). In breast cancer, altered \u003cem\u003eLYN\u003c/em\u003e expression/mutation and rewired RTK\u0026ndash;integrin\u0026ndash;EMT networks plausibly promote CNS tropism\u0026mdash;from intravasation and circulatory survival to endothelial adhesion and BBB transmigration (\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e).\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003e1.2. Molecular Architecture of \u003cem\u003eLYN\u003c/em\u003e\u003c/h2\u003e\u003cp\u003eLike other Src family kinases (SFKs), \u003cem\u003eLYN\u003c/em\u003e comprises modular domains that govern membrane targeting, autoinhibition, and stimulus-coupled activation:\u003c/p\u003e\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003eN-terminus / SH4-Unique: An N-terminal glycine is myristoylated to enable membrane association; palmitoylation and basic residues in the Unique region further stabilize interactions with specific membrane microdomains and organelles. This region dictates subcellular localization-and thus receptor/scaffold access-and alternative splicing fine-tunes these effects (\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e).\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eSH3 domain: Binds proline-rich motifs in partners and in \u003cem\u003eLYN\u003c/em\u003e's own linker, clamping the kinase in a closed, autoinhibited state. High-affinity exogenous ligands can outcompete the intramolecular interaction to promote opening and priming (\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e).\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eSH2 domain: Recognizes phosphotyrosine motifs. In autoinhibition, it engages the phosphorylated C-terminal regulatory tyrosine, reinforcing the closed conformation; binding to external pTyr ligands displaces this contact, activates the kinase, and recruits \u003cem\u003eLYN\u003c/em\u003e to new complexes (\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e).\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eSH2-kinase linker: A flexible, proline-rich segment that transmits conformational signals from the SH3/SH2 \"regulatory head\" to the catalytic core, acting as a mechanical gate that shifts the closed-open equilibrium and modulates catalytic efficiency (\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e).\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eKinase domain: The bilobal catalytic core contains the activation loop, whose phosphorylation state and conformation control substrate access and turnover. Subtle changes in loop dynamics-via mutation or scaffolding-can markedly alter activity and specificity (\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e).\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eC-terminal regulatory tyrosine / CSK axis: CSK-mediated phosphorylation of the tail tyrosine promotes SH2 docking and autoinhibition; dephosphorylation by receptor-proximal phosphatases releases the clamp. The CSK-phosphatase balance thus sets a tunable rheostat on \u003cem\u003eLYN\u003c/em\u003e activity (\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e).\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e\u003cp\u003eCollectively, these modules create a conformational switch that integrates lipid targeting, intramolecular clamps, and catalytic licensing to couple localization to signaling (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). In epithelial and breast cancer settings, perturbations in membrane targeting, regulatory interfaces, or the kinase core can yield distinct phenotypes-altered focal-adhesion dynamics, EMT programs, and RTK/integrin responses-motivating tests of whether variant classes (e.g., truncations vs. interface/targeting missense changes) associate with CNS metastasis (\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan additionalcitationids=\"CR14\" citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec4\" class=\"Section2\"\u003e\u003ch2\u003e1.3. \u003cem\u003eLYN\u003c/em\u003e Alterations in Breast Cancer\u003c/h2\u003e\u003cp\u003eCopy-number and expression changes: Across cancers, amplification/overexpression of \u003cem\u003eLYN\u003c/em\u003e has been reported more frequently than recurrent hotspot mutations, suggesting dosage-sensitive contributions to signaling output and plasticity (\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eIn breast cancer, point mutations are relatively rare and heterogeneous in position and type (missense, occasional truncations/splice). Most single-nucleotide variants lack definitive functional annotation, and their effects likely depend on domain context (e.g., SH4/Unique vs kinase core) and cellular compartmentalization (\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eFor the majority of variants, oncogenicity remains uncertain and there is no established clinical actionability. Consequently, mapping variants to structural/functional modules and to phenotypes such as CNS metastasis may help prioritize variants for mechanistic study and clinical surveillance. Despite extensive pathway crosstalk with receptor tyrosine kinases and integrins, there are no approved \u003cem\u003eLYN\u003c/em\u003e-directed therapies in breast cancer; early clinical experience with multi-target Src family kinase (SFK) inhibitors has been mixed, highlighting the importance of biomarker-guided strategies rather than unselected use (\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e).\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec5\" class=\"Section2\"\u003e\u003ch2\u003e1.4. CNS Metastasis Biology and the Blood\u0026ndash;Brain Barrier (BBB)\u003c/h2\u003e\u003cp\u003eBrain metastasis proceeds through sequential bottlenecks: (i) local invasion and intravasation, (ii) survival in circulation and vascular arrest within the cerebral microvasculature, (iii) adhesion to brain endothelium and transmigration across the BBB (paracellular and/or transcellular routes), and (iv) adaptation to the neural niche, including crosstalk with astrocytes, microglia, and perivascular cells that governs early survival and outgrowth (\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThe BBB\u0026rsquo;s tight junctions, low transcytosis, and specialized perivascular architecture (endothelial cells, pericytes, astrocyte end-feet) impose stringent barrier and signaling constraints on circulating tumor cells (\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e).\u003c/p\u003e\u003cp\u003e\u003cem\u003eLYN\u003c/em\u003e could influence CNS tropism at multiple nodes:\u003c/p\u003e\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003eOn the tumor cell, by coordinating integrin\u0026ndash;FAK\u0026ndash;SFK signaling for endothelial adhesion, cytoskeletal rearrangement, and diapedesis; by modulating RTK-driven survival during transit; and by altering membrane targeting (via the SH4/Unique region) that dictates receptor proximity.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eIn the microenvironment, \u003cem\u003eLYN\u003c/em\u003e-dependent immune signaling may shape endothelial permeability and perivascular inflammation, indirectly affecting BBB traversal.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eWithin the endothelium, SFK activity contributes to junctional dynamics; perturbations in \u003cem\u003eLYN\u003c/em\u003e-centered pathways could tilt the balance toward passage or retention.\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec6\" class=\"Section2\"\u003e\u003ch2\u003e1.5. Rationale, Objective, and Prespecified Hypotheses\u003c/h2\u003e\u003cp\u003eRationale. Given \u003cem\u003eLYN\u003c/em\u003e\u0026rsquo;s membrane targeting, conformational regulation, and positioning at the crossroads of integrin/FAK and RTK signaling, perturbations in \u003cem\u003eLYN\u003c/em\u003e\u0026mdash;especially those altering regulatory interfaces or the catalytic core\u0026mdash;could modulate key steps of CNS tropism, including endothelial adhesion and BBB transmigration.\u003c/p\u003e\u003cp\u003eThis study aims to investigate whether the class of \u003cem\u003eLYN\u003c/em\u003e mutations (truncating vs. missense) and their domain localization are associated with CNS metastasis in breast cancer. Based on the structural logic of \u003cem\u003eLYN\u003c/em\u003e, we hypothesize that truncating mutations\u0026mdash;by disrupting regulatory/tail elements\u0026mdash;and N-terminal SH4/Unique alterations\u0026mdash;by modifying membrane targeting\u0026mdash;are enriched in CNS-positive cases, consistent with mechanisms facilitating endothelial adhesion, BBB transmigration, and neural niche adaptation.\u003c/p\u003e\u003cp\u003eThis work provides a domain-resolved view of somatic \u003cem\u003eLYN\u003c/em\u003e alterations in relation to CNS involvement in breast cancer, linking structural modules to metastatic phenotypes. By pooling harmonized, multi-cohort data and prespecifying LYN domain- and variant class\u0026ndash;based hypotheses, we aim to generate mechanistically grounded, testable leads that prioritize biomarker development and guide functional validation.\u003c/p\u003e\u003c/div\u003e"},{"header":"2. Methods","content":"\u003cp\u003e\u003cstrong\u003e2.1. Study Design and Cohort\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe conducted a retrospective analysis of publicly available breast cancer cohorts accessed via cBioPortal (20-22) (download month: September 2025). All included data are de‑identified and publicly released by the contributing studies; no direct interaction with human subjects occurred.\u003c/p\u003e\n\u003cp\u003eWe conducted a pooled, retrospective cohort study by harmonizing breast cancer genomics and clinical annotations across 29 independent studies, yielding an analytic cohort of 5,947 unique tumors. Harmonization steps included: (i) de-duplication using stable patient/sample identifiers across portals, (ii) standardization of variant formats and clinical fields, and (iii) exclusion of records lacking essential identifiers or \u003cem\u003eLYN\u003c/em\u003e coverage. Inclusion criteria were histologically confirmed breast cancer, next-generation sequencing with ascertainable \u003cem\u003eLYN\u003c/em\u003e status (wild-type vs mutant), and availability of metastatic site fields. Reporting follows STROBE recommendations for observational studies.\u003c/p\u003e\n\u003cp\u003eExclusion criteria: duplicate or discordant records, samples failing basic quality checks, or missing essential identifiers.\u003c/p\u003e\n\u003cp\u003eEthical conduct followed local regulations and the Declaration of Helsinki. Reporting adheres to STROBE recommendations for observational studies.\u003c/p\u003e\n\u003cp\u003eThis process encompassed 33 studies totaling n = 17,605 tumors; 29 contributed NST tumors with LYN coverage to the analytic set (n=5,947). Of these, n = 16,881 had somatic mutation profiling; applying the NST diagnosis restriction yielded n = 14,050. Within the NST set, n = 5,947 tumors (from 29 studies) had \u003cem\u003eLYN\u003c/em\u003e sequenced and formed the analytic denominator for \u003cem\u003eLYN\u003c/em\u003e‑status analyses. Where a study contributed multiple samples per patient, we prioritized the earliest profiled metastatic or primary specimen per patient identifier when feasible (study‑specific rules documented in the analysis notebook). Cross‑study duplicate patients were assessed by hashed combinations of sex/age/diagnosis year and removed when confidently identified. Study-level contributions were:\u003c/p\u003e\n\u003cp\u003e• breast_msk_2025 (n=1,566)\u003c/p\u003e\n\u003cp\u003e• brca_tcga (n=980)\u003c/p\u003e\n\u003cp\u003e• breast_ink4_msk_2021 (n=639)\u003c/p\u003e\n\u003cp\u003e• breast_msk_2018 (n=415)\u003c/p\u003e\n\u003cp\u003e• brca_msk_2025 (n=404)\u003c/p\u003e\n\u003cp\u003e• brca_mbcproject_2022 (n=379)\u003c/p\u003e\n\u003cp\u003e• brca_igr_2015 (n=216)\u003c/p\u003e\n\u003cp\u003e• brca_smc_2018 (n=186)\u003c/p\u003e\n\u003cp\u003e• brca_aurora_2023 (n=134)\u003c/p\u003e\n\u003cp\u003e• breast_cptac_gdc (n=124)\u003c/p\u003e\n\u003cp\u003e• brca_cptac_2020 (n=122)\u003c/p\u003e\n\u003cp\u003e• brca_bccrc_xenograft_2014 (n=116)\u003c/p\u003e\n\u003cp\u003e• brca_broad (n=103)\u003c/p\u003e\n\u003cp\u003e• brca_sanger (n=100)\u003c/p\u003e\n\u003cp\u003e• brca_fuscc_2020 (n=69)\u003c/p\u003e\n\u003cp\u003e• brca_bccrc (n=65)\u003c/p\u003e\n\u003cp\u003e• brca_mapk_hp_msk_2021 (n=61)\u003c/p\u003e\n\u003cp\u003e• brca_mbcproject_wagle_2017 (n=61)\u003c/p\u003e\n\u003cp\u003e• brca_pareja_msk_2020 (n=60)\u003c/p\u003e\n\u003cp\u003e• brca_dfci_2020 (n=59)\u003c/p\u003e\n\u003cp\u003e• bfn_duke_nus_2015 (n=22)\u003c/p\u003e\n\u003cp\u003e• mbc_msk_2021 (n=18)\u003c/p\u003e\n\u003cp\u003e• ilc_msk_2023 (n=14)\u003c/p\u003e\n\u003cp\u003e• acbc_mskcc_2015 (n=12)\u003c/p\u003e\n\u003cp\u003e• breast_alpelisib_2020 (n=10)\u003c/p\u003e\n\u003cp\u003e• brca_jup_msk_2020 (n=5)\u003c/p\u003e\n\u003cp\u003e• brca_hta9_htan_2022 (n=5)\u003c/p\u003e\n\u003cp\u003e• brca_tcga_gdc (n=2)\u003c/p\u003e\n\u003cp\u003e• brca_tcga_pan_can_atlas_2018 (n=1).\u003cstrong\u003e\u003cbr\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eDirect study links are provided in the Declarations (Availability of data and materials).\u003c/p\u003e\n\u003cp\u003eMolecular subtype and missing data policy. ER/PR/HER2 molecular subtype variables were not uniformly captured across contributing studies; HER2 status in particular was missing in the majority of tumors and often inconsistently reported between primary and metastatic specimens. We therefore prespecified that subtype variables would not be included as covariates in the primary pooled analyses, and no imputation would be attempted. Because conditioning on a partially observed determinant of CNS risk (e.g., HER2) can introduce selection bias, we also refrained from restricting the analysis to complete-case subsets.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.2. Endpoints and Variable Definitions\u003c/strong\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe primary endpoint was central nervous system (CNS) metastasis at any time (yes/no), harmonized across studies using a controlled vocabulary. Metastatic sites were harmonized by mapping free-text or coded fields to a controlled vocabulary. We considered entries such as “brain”, “cerebellum”, “dura”, “occipital”, and “leptomeningeal/CSF” as CNS-positive (case-insensitive substring matching). Site fields absent were retained as missing; no imputation was performed.\u003c/p\u003e\n\u003cp\u003eDates of CNS involvement were not systematically captured across contributing studies; therefore, baseline versus incident CNS events could not be distinguished. Time-to-event analyses (e.g., Cox or Fine–Gray) were not feasible, and we prespecified a binary “ever-CNS” endpoint without imputation of missing dates. Reported associations thus reflect odds of ever having CNS involvement in the pooled dataset rather than hazards or cumulative incidence over time.\u003c/p\u003e\n\u003cp\u003eImaging intensity and follow-up schedules likely varied across studies; we did not adjust for follow-up time because event dates were unavailable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.3. Variant Ascertainment and Patient-Level Classification\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSomatic \u003cem\u003eLYN\u003c/em\u003e variants (coding SNVs/indels) were collected from each study’s mutation tables, standardized, and collapsed to patient-level classes. Where multiple \u003cem\u003eLYN\u003c/em\u003e variants occurred in the same patient, all were retained at the variant level and then collapsed to a patient-level class:\u003c/p\u003e\n\u003cul\u003e\n \u003cli\u003eTruncating: any nonsense, frameshift, or canonical splice-site event present (patient labeled \"truncating\" irrespective of co-occurring missense variants).\u003c/li\u003e\n \u003cli\u003eMissense-only: one or more missense variants with no truncating/splice events.\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003eComposite strings (e.g., multiple amino-acid changes) were parsed into constituent events; classification defaulted to truncating if any component met truncation criteria.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.4. Domain Mapping\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFor each variant, we extracted the codon position (when determinable) and mapped it to canonical Src-family modules: SH4/Unique (N-terminus), SH3, SH2, SH2-kinase linker, and kinase domain (activation loop noted as a subregion within the kinase domain). Splice variants without a resolvable codon number were labeled splice; frameshifts were assigned to the position of the first impacted codon. Variants lacking a reliable position were labeled unknown for domain analyses.\u003c/p\u003e\n\u003cp\u003eMulti-hit policy. We prespecified that if a tumor harbored variants mapping to multiple domains, it would be collapsed to the most proximal domain. In this dataset, no tumors carried variants in more than one domain; thus, domain collapsing was not required.\u003c/p\u003e\n\u003cp\u003eUnknown mapping. Variants without a resolvable protein position (e.g., splice events without a defined codon, ambiguous frameshifts) were classified as “unknown domain” and excluded from domain-level tests but retained in overall LYN-mutant analyses.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.5. Statistical Analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFor binary comparisons, we used two-sided Fisher's exact tests and reported odds ratios (ORs) with exact 95% confidence intervals. Where any cell count was zero, we additionally reported the Haldane-Anscombe corrected OR for interpretability. Alongside pooled Fisher’s tests, we considered study-stratified Mantel–Haenszel and random-effects pooling, but did not report these due to sparse per-study cells and unstable estimates.\u003c/p\u003e\n\u003cp\u003eGiven small sample sizes, we fit a Firth bias-reduced logistic model with CNS(+) as the dependent variable and LYN mutation class (truncating vs missense) as the covariate.\u003c/p\u003e\n\u003cp\u003eEach patient was mapped to domain presence (yes/no) for SH4/Unique, SH3, SH2, linker, and kinase domains, and per-domain Fisher's tests were conducted against CNS status. Domain-presence screens were considered hypothesis-generating; all reported p-values were Benjamini-Hochberg FDR-adjusted.\u003c/p\u003e\n\u003cp\u003eAnalyses were performed in R (version 4.3.2) using stats::fisher.test and logistf packages. All tests were two-sided with α = 0.05 unless stated otherwise. Effect sizes and their uncertainty were prioritized over thresholded significance.\u003c/p\u003e\n\u003cp\u003eTo account for between-study heterogeneity in sampling and CNS ascertainment, where estimable regression models included a study fixed effect (dummy-coded). For binary comparisons, alongside pooled Fisher’s exact tests, we considered MH and RE pooling, but did not report due to sparse per-study cells. Where appropriate, we performed random-effects inverse-variance meta-analysis of study-specific log-ORs. The prespecified Firth bias-reduced logistic model was extended to include the study effect. Two-sided α=0.05 unless stated.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.5.1. P-Values and Confidence Intervals\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTwo-sided p-values for 2 x 2 comparisons were obtained using conditional Fisher's exact test. Alongside pooled Fisher’s tests, we considered Mantel–Haenszel and random-effects pooling, but did not report these due to sparse per-study cells and unstable estimates. For the Firth bias-reduced logistic model (CNS(+) ~ truncating), we report profile penalized-likelihood 95% CIs for coefficients (reported as OR = e\u003csup\u003eβ\u003c/sup\u003e) and penalized likelihood-ratio test (LRT) p-values. Wald intervals were not used.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.5.2. Sensitivity Analyses\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSensitivity analyses included rare-variant thresholding (excluding singleton somatic \u003cem\u003eLYN\u003c/em\u003e variants or those below a pre-defined frequency cut-off). For each scenario we re-fit Fisher and Firth models and report OR (95% CI) and p-values, emphasizing effect direction and precision rather than thresholded significance.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.6. Variant-Level Clinical Annotation\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFor each observed \u003cem\u003eLYN\u003c/em\u003e variant, we extracted OncoKB (23, 24) (accessed 2025-09-08)\u0026nbsp;annotations and integrated into the analysis workbook: oncogenicity statements (e.g., unknown oncogenic effect, likely oncogenic) and therapy notes (presence/absence of FDA-approved or compendium-listed options relevant to the alteration and disease context).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.7. Functional enrichment\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe queried STRING (Homo sapiens; accessed 2025-09-11) seeding \u003cem\u003eLYN\u003c/em\u003e with active sources: experiments, databases, co-expression; minimum required interaction score ≥ 0.70. Over-representation was tested against a genome-wide background with Benjamini–Hochberg FDR. We display top non-redundant terms with q \u0026lt; 0.05.\u003c/p\u003e"},{"header":"3. Results","content":"\u003cp\u003e\u003cstrong\u003e3.1 Cohort Overview and Ascertainment of \u003cem\u003eLYN\u003c/em\u003e Status\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe pooled analytic cohort comprised 5,947 unique breast cancer samples from 29 independent studies after de-duplication and quality control. \u003cem\u003eLYN\u003c/em\u003e status was determinable in samples where panel coverage included \u003cem\u003eLYN\u003c/em\u003e; CNS metastasis was harmonized using a controlled vocabulary across studies (Figure 3).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e3.2 Association Between \u003cem\u003eLYN\u003c/em\u003e Mutation (Present vs WT) and CNS Metastasis\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIn the full dataset (n = 5,947), CNS metastasis occurred in 5/46 \u003cem\u003eLYN\u003c/em\u003e-mutated tumors (10.9%) versus 110/5,901 \u003cem\u003eLYN\u003c/em\u003e wild-type tumors (1.9%), yielding an odds ratio = 6.42 (95% CI, 2.49\u0026ndash;16.56) with a two-sided Fisher\u0026rsquo;s exact p = 0.0018. This suggests a statistically significant excess of CNS involvement among \u003cem\u003eLYN\u003c/em\u003e-mutant cases (Figure 4).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e3.3 Domain-Resolved Distribution of \u003cem\u003eLYN\u003c/em\u003e Mutations by CNS Status\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eOverall heterogeneity across domains was significant (\u0026chi;\u0026sup2; p \u0026asymp; 0.014), indicating that the location of the \u003cem\u003eLYN\u003c/em\u003e alteration may influence CNS risk. One-versus-rest contrasts (Benjamini\u0026ndash;Hochberg FDR control)(Figure 5):\u003c/p\u003e\n\u003cul\u003e\n \u003cli\u003eUNIQUE/SH4 (N-terminus): 3/3 in-domain cases were CNS-positive (100.0%), versus 2/16 in the remainder (12.5%). The 2\u0026times;2 table contains a zero cell, yielding an infinite exact odds ratio; using a Haldane\u0026ndash;Anscombe continuity correction for interpretability gives OR \u0026asymp; 46.2. Unadjusted p \u0026asymp; 0.010; FDR q \u0026asymp; 0.052. For zero-cell tables, we report qualitative inference from exact testing and, for interpretability, continuity-corrected odds ratios (Haldane\u0026ndash;Anscombe). Multiple testing for domain-level one-versus-rest contrasts used BH-FDR (UNIQUE q \u0026asymp; 0.052; others not significant).\u003c/li\u003e\n \u003cli\u003eSH2: 0/4 in-domain CNS-positive (0.0%) versus 5/15 in the remainder (33.3%); Fisher p \u0026asymp; 0.53. Continuity-corrected OR \u0026asymp; 0.21.\u003c/li\u003e\n \u003cli\u003eSH3: 0/1 in-domain CNS-positive (0.0%) versus 5/18 in the remainder (27.8%); Fisher p \u0026asymp; 1.00. Continuity-corrected OR \u0026asymp; 0.82.\u003c/li\u003e\n \u003cli\u003eKinase domain: 2/11 in-domain CNS-positive (18.2%) versus 3/8 in the remainder (37.5%); Fisher p \u0026asymp; 0.60. OR \u0026asymp; 0.37.\u003c/li\u003e\n \u003cli\u003eSH2\u0026ndash;kinase linker: No in-domain observations; not evaluable.\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003e\u003cstrong\u003e3.4 Mutation Class (Truncating vs Missense) by CNS Status\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAmong \u003cem\u003eLYN\u003c/em\u003e-mutant cases, CNS metastasis was observed in 3/38 (7.9%) with missense variants and 2/7 (28.6%) with truncating variants. A direct Fisher\u0026rsquo;s exact comparison of these class-specific rates did not reach significance (two-sided p = 0.166) (Figure 6).\u003c/p\u003e\n\u003cp\u003eFramed alternatively as the proportion of truncating variants among CNS-positive vs CNS-negative cases, the truncating share was 2/5 (40.0%) vs 6/44 (13.6%), yielding OR = 4.22 (exact 95% CI, 0.58\u0026ndash;30.75; two-sided Fisher p = 0.182). Thus, although the point estimate exceeds 1, the confidence interval is wide and includes the null under the available sample size.\u003c/p\u003e\n\u003cp\u003eIn a prespecified Firth logistic model for CNS(+) ~ truncating (vs missense) restricted to \u003cem\u003eLYN\u003c/em\u003e-mutant tumors, estimates were directionally consistent with the Fisher analysis but non-significant; profile penalized-likelihood 95% CIs were wide, reflecting sparse counts (particularly truncating, n = 7).\u003c/p\u003e"},{"header":"4. Discussion","content":"\u003cdiv id=\"Sec23\" class=\"Section2\"\u003e\u003ch2\u003e4.1. Principal findings\u003c/h2\u003e\u003cp\u003eIn this pooled, retrospectively harmonized multi-study cohort, somatic \u003cem\u003eLYN\u003c/em\u003e mutation was associated with a higher frequency of CNS metastasis compared with \u003cem\u003eLYN\u003c/em\u003e wild type: 10.9% (5/46) versus 1.9% (110/5,901), respectively (OR\u0026thinsp;=\u0026thinsp;6.42; 95% CI, 2.49\u0026ndash;16.56; two-sided Fisher p\u0026thinsp;=\u0026thinsp;0.0018; n\u0026thinsp;=\u0026thinsp;5,947). Within the \u003cem\u003eLYN\u003c/em\u003e-mutant subset with resolvable domain mapping, domain distributions differed by CNS status (omnibus χ\u0026sup2; p\u0026thinsp;\u0026asymp;\u0026thinsp;0.014). In one-versus-rest contrasts, the SH4/Unique (N-terminus) was represented among CNS-positive cases (3/3) but not among CNS-negative entries for that domain, whereas other modules (SH2, SH3, kinase) did not show enrichment after multiplicity control (UNIQUE unadjusted p\u0026thinsp;\u0026asymp;\u0026thinsp;0.010; FDR q\u0026thinsp;\u0026asymp;\u0026thinsp;0.052; Haldane\u0026ndash;Anscombe OR\u0026thinsp;\u0026asymp;\u0026thinsp;46.2, acknowledging a zero cell and n\u0026thinsp;=\u0026thinsp;3). By mutation class, truncating events were proportionally higher among CNS-positive than CNS-negative \u003cem\u003eLYN\u003c/em\u003e mutants (40.0% vs 13.6%), but this comparison did not reach statistical significance (OR\u0026thinsp;=\u0026thinsp;4.22; exact 95% CI, 0.58\u0026ndash;30.75; Fisher p\u0026thinsp;=\u0026thinsp;0.182). Collectively, these observations support a domain-aware pattern relating \u003cem\u003eLYN\u003c/em\u003e variation to CNS involvement, with an N-terminal signal that is striking but sample-limited and therefore hypothesis-generating.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec24\" class=\"Section2\"\u003e\u003ch2\u003e4.2. Biological context and plausibility\u003c/h2\u003e\u003cp\u003eThe SH4/Unique segment of Src-family kinases governs membrane targeting via N-terminal lipidation and basic residue interactions with specific microdomains. An enrichment signal at this module\u0026mdash;albeit based on very small counts\u0026mdash;is biologically coherent with steps requisite for CNS spread: endothelial adhesion, BBB transmigration, and early perivascular survival, processes that depend on precise spatial coupling of integrins/FAK, RTKs, and cytoskeletal machinery. By contrast, the kinase domain was represented across CNS strata without clear differential enrichment, suggesting that localization/context\u0026mdash;rather than catalytic-core involvement alone\u0026mdash;may distinguish CNS behavior for \u003cem\u003eLYN\u003c/em\u003e-mutant tumors in this dataset. These data motivate functional tests of how N-terminal perturbations rewire receptor\u0026ndash;adhesion complexes and barrier interactions in brain endothelium models.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec25\" class=\"Section2\"\u003e\u003ch2\u003e4.3. Interpreting the mutation-class signal\u003c/h2\u003e\u003cp\u003eThe truncating-versus-missense comparison yielded point estimates\u0026thinsp;\u0026gt;\u0026thinsp;1 but non-significant p values under exact testing (p\u0026thinsp;=\u0026thinsp;0.166 for class-specific rates; p\u0026thinsp;=\u0026thinsp;0.182 for the truncating share analysis), with wide CIs reflecting sparse truncating counts (n\u0026thinsp;=\u0026thinsp;7) and only five CNS-positive \u003cem\u003eLYN\u003c/em\u003e-mutant cases overall. These results are compatible with either a true but imprecise effect or no effect; as such, they should not be over-interpreted absent replication or larger pooled analyses and, ideally, orthogonal functional evidence.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec26\" class=\"Section2\"\u003e\u003ch2\u003e4.4. Mechanistic considerations: \u003cem\u003eLYN\u003c/em\u003e, CNS tropism, and the tumor microenvironment\u003c/h2\u003e\u003cp\u003e\u003cem\u003eLYN\u003c/em\u003e sits at a nodal position connecting receptor tyrosine kinases, integrins, immune receptors and cytoskeletal scaffolds. Perturbations of this hub\u0026mdash;whether by mutation, truncation, or domain-specific rewiring\u0026mdash;could plausibly shift adhesion, motility and survival programs in ways that favor CNS dissemination. Beyond tumor-intrinsic effects, \u003cem\u003eLYN\u003c/em\u003e also operates in endothelial and immune compartments, suggesting multi-cellular routes to brain tropism (\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eA first mechanistic axis concerns membrane targeting and nanoscale localization. The SH4/Unique N-terminus controls myristoylation/palmitoylation-dependent membrane association and partitioning into lipid microdomains; even small changes here can redirect the kinase to distinct receptor/scaffold neighborhoods. In principle, N-terminal alterations could bias \u003cem\u003eLYN\u003c/em\u003e toward complexes that enhance endothelial adhesion and barrier engagement, or impair autoinhibitory clamping, thereby lowering the activation threshold at the cell periphery (\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e, \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e). Such relocalization would be expected to remodel proximal phosphorylation of focal-adhesion adaptors and junctional proteins.\u003c/p\u003e\u003cp\u003eSecond, adhesion\u0026ndash;cytoskeleton coupling provides a convergent route to motility and diapedesis. \u003cem\u003eLYN\u003c/em\u003e phosphorylates focal-adhesion components (e.g., BCAR1/CAS, NEDD9/HEF1) and interfaces with integrin/FAK signaling to accelerate adhesion turnover and directional migration (\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e). In parallel, cross-talk with ERBB/EGFR and downstream PI3K\u0026ndash;AKT and MAPK cascades can increase survival of intravasating or circulating cells, potentially enlarging the pool of cells that can reach and interrogate the brain endothelium (\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThird, endothelial and barrier biology may link \u003cem\u003eLYN\u003c/em\u003e to the blood\u0026ndash;brain barrier (BBB). \u003cem\u003eLYN\u003c/em\u003e activity has been implicated in regulating endothelial activation, leukocyte adhesion and transendothelial migration, processes that mirror the steps required for tumor cell extravasation (\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e). If \u003cem\u003eLYN\u003c/em\u003e-dependent signaling modulates endothelial permeability or the presentation of adhesion molecules under inflammatory cues, tumor cells with LYN pathway perturbations could gain context-dependent advantages at the BBB interface (\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eFourth, immune regulation offers a microenvironmental bridge. \u003cem\u003eLYN\u003c/em\u003e can act as a negative regulator by phosphorylating ITIM-containing receptors and recruiting phosphatases (SHP-1/SHP-2/SHIP-1), damping immune activation in B cells and myeloid lineages while still permitting context-specific activation (\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e). In a tumor setting, such tuning might blunt anti-tumor immunity locally or systemically, alter neutrophil/monocyte trafficking, and reshape cytokine milieus that influence BBB function and niche formation (\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e). These effects need not be restricted to tumor cells; host leukocytes with \u003cem\u003eLYN\u003c/em\u003e-dependent signaling could indirectly facilitate metastatic seeding.\u003c/p\u003e\u003cp\u003eFifth, chemokine and growth-factor axes downstream of \u003cem\u003eLYN\u003c/em\u003e (e.g., CXCR4, EPOR, KIT) point to additional, testable circuits. CXCL12\u0026ndash;CXCR4 gradients are operative at vascular and neural interfaces; \u003cem\u003eLYN\u003c/em\u003e-mediated propagation of such signals could coordinate chemotaxis with adhesion mechanics. Likewise, \u003cem\u003eLYN\u003c/em\u003e\u0026rsquo;s links to platelet and erythroid signaling raise the possibility that platelet cloaking or altered rheology contributes to intravascular survival and endothelial docking, although these ideas remain speculative without matched functional data.\u003c/p\u003e\u003cp\u003eTogether, these lines of reasoning generate concrete predictions. If N-terminal/domain context matters for CNS tropism, \u003cem\u003eLYN\u003c/em\u003e-mutant tumors that seed the CNS should show (i) altered subcellular localization and membrane microdomain occupancy; (ii) rewired phosphoproteomic signatures at focal adhesions and junctions; (iii) enhanced adhesion/diapedesis across human brain microvascular endothelial monolayers under shear; and (iv) microenvironmental differences in endothelial activation and immune infiltration. These can be evaluated in isogenic systems expressing representative SH4/Unique versus kinase-core variants, using live-cell imaging, microfluidic BBB models, and unbiased phospho-proteomics; in clinical material, spatial IHC or transcriptomic deconvolution (B-cell and myeloid signatures, adhesion/chemokine programs) could probe whether the \u003cem\u003eLYN\u003c/em\u003e\u0026ndash;CNS signal co-travels with microenvironmental features.\u003c/p\u003e\u003cp\u003eFinally, the therapeutic implications are purposely cautious. Pan-SFK inhibition has produced mixed results and may be limited by BBB penetration and pathway redundancy; conversely, domain-driven rewiring suggests combinations\u0026mdash;e.g., SRC-family inhibitors with integrin/FAK blockade or anti-inflammatory modulators\u0026mdash;could be more effective in biologically defined subsets (\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e, \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e). At present, our findings are hypothesis-generating; establishing independence from canonical clinical drivers and embedding the \u003cem\u003eLYN\u003c/em\u003e signal within a microenvironmental framework will require subtype-balanced, time-resolved datasets and mechanistic validation across the tumor\u0026ndash;endothelium\u0026ndash;immune triad.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec27\" class=\"Section2\"\u003e\u003ch2\u003e4.4. Methodological strengths\u003c/h2\u003e\u003cp\u003eThis work adheres to several practices that increase inferential robustness for rare alterations and sparse endpoints:\u003c/p\u003e\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003ePrespecified endpoint harmonization using a controlled CNS vocabulary with no imputation for missing site fields;\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eExact procedures appropriate to small cells (two-sided Fisher tests, exact/mid-P intervals where feasible) and BH-FDR control for domain screens;\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eA priori domain mapping and patient-level collapsing rules;\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eMulti-study pooling with de-duplication across patient/sample identifiers to reduce double counting.\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec28\" class=\"Section2\"\u003e\u003ch2\u003e4.5. Limitations\u003c/h2\u003e\u003cp\u003eImportant caveats temper interpretation. The retrospective design risks selection and information bias from clinically sequenced cohorts and free-text metastasis fields (despite normalization). The \u003cem\u003eLYN\u003c/em\u003e-mutant denominator is small (n\u0026thinsp;=\u0026thinsp;46; CNS-positive n\u0026thinsp;=\u0026thinsp;5), yielding imprecise estimates with wide CIs and limiting multivariable adjustment. Key clinical/genomic covariates (subtype, treatment, stage, co-mutations, copy number/expression, TMB) were incompletely available, leaving residual confounding or effect modification. CNS event timing (baseline vs follow-up) was inconsistently captured, precluding temporal inference. Finally, heterogeneity in panel coverage and site capture across studies may introduce between-study variability only partially addressed by pooled exact analyses.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec29\" class=\"Section2\"\u003e\u003ch2\u003e4.6. Implications and next steps\u003c/h2\u003e\u003cp\u003eThese findings elevate domain context as a key lens for assessing LYN variation and CNS risk. Next steps:\u003c/p\u003e\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003eExternal validation: Well-annotated cohorts with harmonized CNS fields and sufficient \u003cem\u003eLYN\u003c/em\u003e-mut/CNS\u0026thinsp;+\u0026thinsp;cases to refine estimates and enable covariate-adjusted models (e.g., subtype-adjusted Firth logistic; study-stratified or random-effects meta-analysis).\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eTemporal curation: Time-stamp CNS events to separate baseline from incident metastases and test lead-lag relationships with LYN status.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eMechanistic studies: Isogenic models comparing SH4/Unique vs kinase-core/regulatory-head perturbations, probing membrane targeting, endothelial adhesion, and diapedesis in reductionist and microphysiological BBB systems.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eSignaling/interactome profiling: Determine whether N-terminal alterations reprogram RTK-integrin coupling or scaffold composition to favor CNS tropism.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eClinical pipelines: Domain-aware curation to standardize future aggregation/meta-analysis and flag putatively high-risk patterns for prospective surveillance.\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e\u003c/div\u003e"},{"header":"5. Conclusion","content":"\u003cp\u003eIn a harmonized, multi-study breast cancer cohort, any somatic \u003cem\u003eLYN\u003c/em\u003e mutation was associated with increased odds of CNS metastasis (10.9% vs 1.9%; OR\u0026thinsp;=\u0026thinsp;6.42; 95% CI, 2.49\u0026ndash;16.56; p\u0026thinsp;=\u0026thinsp;0.0018). Among \u003cem\u003eLYN\u003c/em\u003e-mutant tumors, domain distributions differed by CNS status, with a small-sample enrichment signal at the SH4/Unique N-terminus (FDR-borderline), while other modules did not show significant enrichment. The truncating-versus-missense comparison was directionally positive but statistically non-significant under sparse counts, and should be viewed as exploratory. Overall, the data elevate domain context\u0026mdash;particularly N-terminal membrane-targeting logic\u0026mdash;as a mechanistically plausible axis for future validation and functional interrogation. Given the limited power and heterogeneous annotations, immediate therapeutic implications are premature, especially considering the challenges of CNS drug delivery and the mixed performance of unselected SFK inhibition. Prospective validation in larger, uniformly annotated datasets, coupled with domain-informed experiments centered on membrane localization and BBB traversal, will be critical to determine whether \u003cem\u003eLYN\u003c/em\u003e domain patterns can inform risk modeling or biomarker development in CNS involvement of breast cancer.\u003c/p\u003e\u003cp\u003eFindings are hypothesis-generating and require independent validation in larger, uniformly annotated cohorts before any clinical application.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cp\u003eCNS, BBB, SFK, RTK, EMT, OR, CI, FDR, IRB, NST, SNV.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate:\u003c/strong\u003e This study analyzed only publicly available, de-identified data aggregated from multiple studies and portals. There was no interaction with human participants and no access to identifiable private information. In accordance with applicable regulations and institutional policies, this secondary analysis of de-identified public data is not considered human subjects research; institutional review board oversight and informed consent were not required. Reporting follows STROBE.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication:\u003c/strong\u003e Not applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and materials:\u003c/strong\u003e All datasets analyzed in this study are publicly accessible from cBioPortal for Cancer Genomics. Direct study links and identifiers are provided below. We also used OncoKB for variant-level annotations and STRING for enrichment analyses; access dates are indicated.\u003c/p\u003e\n\u003cul\u003e\n \u003cli\u003ecBioPortal for Cancer Genomics — breast_msk_2025:\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003ehttps://www.cbioportal.org/study/summary?id=breast_msk_2025\u003c/p\u003e\n\u003cul\u003e\n \u003cli\u003ecBioPortal for Cancer Genomics — brca_tcga:\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003ehttps://www.cbioportal.org/study/summary?id=brca_tcga\u003c/p\u003e\n\u003cul\u003e\n \u003cli\u003ecBioPortal for Cancer Genomics — breast_ink4_msk_2021:\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003ehttps://www.cbioportal.org/study/summary?id=breast_ink4_msk_2021\u003c/p\u003e\n\u003cul\u003e\n \u003cli\u003ecBioPortal for Cancer Genomics — breast_msk_2018:\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003ehttps://www.cbioportal.org/study/summary?id=breast_msk_2018\u003c/p\u003e\n\u003cul\u003e\n \u003cli\u003ecBioPortal for Cancer Genomics — brca_msk_2025:\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003ehttps://www.cbioportal.org/study/summary?id=brca_msk_2025\u003c/p\u003e\n\u003cul\u003e\n \u003cli\u003ecBioPortal for Cancer Genomics — brca_mbcproject_2022:\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003ehttps://www.cbioportal.org/study/summary?id=brca_mbcproject_2022\u003c/p\u003e\n\u003cul\u003e\n \u003cli\u003ecBioPortal for Cancer Genomics — brca_igr_2015:\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003ehttps://www.cbioportal.org/study/summary?id=brca_igr_2015\u003c/p\u003e\n\u003cul\u003e\n \u003cli\u003ecBioPortal for Cancer Genomics — brca_smc_2018:\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003ehttps://www.cbioportal.org/study/summary?id=brca_smc_2018\u003c/p\u003e\n\u003cul\u003e\n \u003cli\u003ecBioPortal for Cancer Genomics — brca_aurora_2023:\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003ehttps://www.cbioportal.org/study/summary?id=brca_aurora_2023\u003c/p\u003e\n\u003cul\u003e\n \u003cli\u003ecBioPortal for Cancer Genomics — breast_cptac_gdc:\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003ehttps://www.cbioportal.org/study/summary?id=breast_cptac_gdc\u003c/p\u003e\n\u003cul\u003e\n \u003cli\u003ecBioPortal for Cancer Genomics — brca_cptac_2020:\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003ehttps://www.cbioportal.org/study/summary?id=brca_cptac_2020\u003c/p\u003e\n\u003cul\u003e\n \u003cli\u003ecBioPortal for Cancer Genomics — brca_bccrc_xenograft_2014:\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003ehttps://www.cbioportal.org/study/summary?id=brca_bccrc_xenograft_2014\u003c/p\u003e\n\u003cul\u003e\n \u003cli\u003ecBioPortal for Cancer Genomics — brca_broad:\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003ehttps://www.cbioportal.org/study/summary?id=brca_broad\u003c/p\u003e\n\u003cul\u003e\n \u003cli\u003ecBioPortal for Cancer Genomics — brca_sanger:\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003ehttps://www.cbioportal.org/study/summary?id=brca_sanger\u003c/p\u003e\n\u003cul\u003e\n \u003cli\u003ecBioPortal for Cancer Genomics — brca_fuscc_2020 (n=69): https://www.cbioportal.org/study/summary?id=brca_fuscc_2020\u003c/li\u003e\n \u003cli\u003ecBioPortal for Cancer Genomics — brca_bccrc:\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003ehttps://www.cbioportal.org/study/summary?id=brca_bccrc\u003c/p\u003e\n\u003cul\u003e\n \u003cli\u003ecBioPortal for Cancer Genomics — brca_mapk_hp_msk_2021:\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003ehttps://www.cbioportal.org/study/summary?id=brca_mapk_hp_msk_2021\u003c/p\u003e\n\u003cul\u003e\n \u003cli\u003ecBioPortal for Cancer Genomics — brca_mbcproject_wagle_2017:\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003ehttps://www.cbioportal.org/study/summary?id=brca_mbcproject_wagle_2017\u003c/p\u003e\n\u003cul\u003e\n \u003cli\u003ecBioPortal for Cancer Genomics — brca_pareja_msk_2020:\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003ehttps://www.cbioportal.org/study/summary?id=brca_pareja_msk_2020\u003c/p\u003e\n\u003cul\u003e\n \u003cli\u003ecBioPortal for Cancer Genomics — brca_dfci_2020:\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003ehttps://www.cbioportal.org/study/summary?id=brca_dfci_2020\u003c/p\u003e\n\u003cul\u003e\n \u003cli\u003ecBioPortal for Cancer Genomics — bfn_duke_nus_2015:\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003ehttps://www.cbioportal.org/study/summary?id=bfn_duke_nus_2015\u003c/p\u003e\n\u003cul\u003e\n \u003cli\u003ecBioPortal for Cancer Genomics — mbc_msk_2021\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003ehttps://www.cbioportal.org/study/summary?id=mbc_msk_2021\u003c/p\u003e\n\u003cul\u003e\n \u003cli\u003ecBioPortal for Cancer Genomics — ilc_msk_2023: https://www.cbioportal.org/study/summary?id=ilc_msk_2023\u003c/li\u003e\n \u003cli\u003ecBioPortal for Cancer Genomics — acbc_mskcc_2015: https://www.cbioportal.org/study/summary?id=acbc_mskcc_2015\u003c/li\u003e\n \u003cli\u003ecBioPortal for Cancer Genomics — breast_alpelisib_2020: https://www.cbioportal.org/study/summary?id=breast_alpelisib_2020\u003c/li\u003e\n \u003cli\u003ecBioPortal for Cancer Genomics — brca_jup_msk_2020: https://www.cbioportal.org/study/summary?id=brca_jup_msk_2020\u003c/li\u003e\n \u003cli\u003ecBioPortal for Cancer Genomics — brca_hta9_htan_2022: https://www.cbioportal.org/study/summary?id=brca_hta9_htan_2022\u003c/li\u003e\n \u003cli\u003ecBioPortal for Cancer Genomics — brca_tcga_gdc: https://www.cbioportal.org/study/summary?id=brca_tcga_gdc\u003c/li\u003e\n \u003cli\u003ecBioPortal for Cancer Genomics — brca_tcga_pan_can_atlas_2018: https://www.cbioportal.org/study/summary?id=brca_tcga_pan_can_atlas_2018\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003eOther resources:\u003c/p\u003e\n\u003cp\u003eOncoKB — https://www.oncokb.org/ (accessed 2025-09-08)\u003c/p\u003e\n\u003cp\u003eSTRING — https://string-db.org/ (accessed 2025-09-11)\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests:\u003c/strong\u003e The authors declare no competing interests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding:\u003c/strong\u003e No specific funding was received for this work.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors’ contributions:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eE.K.Ç.: Conceptualization; Study design/Methodology; Data curation; Resources; Supervision; Writing—original draft.\u003c/p\u003e\n\u003cp\u003eB.Ç.: Software; Formal analysis/Statistics; Visualization; Writing—original draft.\u003c/p\u003e\n\u003cp\u003eAll authors read and approved the final manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements:\u003c/strong\u003e We gratefully acknowledge the cBioPortal for Cancer Genomics team for providing an open platform to access and analyze cancer genomics data, and the STRING consortium for protein association networks and enrichment tools used in this study. We also thank the contributing investigators, institutions, and—most importantly—the patients—whose data made this research possible. Interpretations and conclusions are solely those of the authors and do not necessarily reflect the views of the data contributors.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eRaghavendra AS, Ibrahim NK. Breast Cancer Brain Metastasis: A Comprehensive Review. JCO Oncol Pract. 2024;20(10):1348-59.\u003c/li\u003e\n\u003cli\u003eStavrou E, Winer EP, Lin NU. How we treat HER2-positive brain metastases. ESMO Open. 2021;6(5):100256.\u003c/li\u003e\n\u003cli\u003eBoire A, Brastianos PK, Garzia L, Valiente M. Brain metastasis. Nat Rev Cancer. 2020;20(1):4-11.\u003c/li\u003e\n\u003cli\u003eSoung YH, Clifford JL, Chung J. Crosstalk between integrin and receptor tyrosine kinase signaling in breast carcinoma progression. BMB Rep. 2010;43(5):311-8.\u003c/li\u003e\n\u003cli\u003eHan J, Zhang G, Welch EJ, Liang Y, Fu J, Vogel SM, et al. A critical role for Lyn kinase in strengthening endothelial integrity and barrier function. Blood. 2013;122(25):4140-9.\u003c/li\u003e\n\u003cli\u003eTornillo G, Knowlson C, Kendrick H, Cooke J, Mirza H, Aurrekoetxea-Rodr\u0026iacute;guez I, et al. Dual Mechanisms of LYN Kinase Dysregulation Drive Aggressive Behavior in Breast Cancer Cells. Cell Rep. 2018;25(13):3674-92.e10.\u003c/li\u003e\n\u003cli\u003eParsons SJ, Parsons JT. Src family kinases, key regulators of signal transduction. Oncogene. 2004;23(48):7906-9.\u003c/li\u003e\n\u003cli\u003eSzklarczyk D, Kirsch R, Koutrouli M, Nastou K, Mehryary F, Hachilif R, et al. The STRING database in 2023: protein-protein association networks and functional enrichment analyses for any sequenced genome of interest. Nucleic Acids Res. 2023;51(D1):D638-d46.\u003c/li\u003e\n\u003cli\u003eChoi YL, Bocanegra M, Kwon MJ, Shin YK, Nam SJ, Yang JH, et al. LYN is a mediator of epithelial-mesenchymal transition and a target of dasatinib in breast cancer. Cancer Res. 2010;70(6):2296-306.\u003c/li\u003e\n\u003cli\u003eSafran M, Rosen N, Twik M, BarShir R, Stein TI, Dahary D, et al. The GeneCards Suite. In: Abugessaisa I, Kasukawa T, editors. Practical Guide to Life Science Databases. Singapore: Springer Nature Singapore; 2021. p. 27-56.\u003c/li\u003e\n\u003cli\u003eDoron H, Pukrop T, Erez N. A Blazing Landscape: Neuroinflammation Shapes Brain Metastasis. Cancer Research. 2019;79(3):423-36.\u003c/li\u003e\n\u003cli\u003eWheeler SE, Morariu EM, Bednash JS, Otte CG, Seethala RR, Chiosea SI, et al. Lyn kinase mediates cell motility and tumor growth in EGFRvIII-expressing head and neck cancer. Clin Cancer Res. 2012;18(10):2850-60.\u003c/li\u003e\n\u003cli\u003eOkada M. Regulation of the SRC family kinases by Csk. Int J Biol Sci. 2012;8(10):1385-97.\u003c/li\u003e\n\u003cli\u003eSicheri F, Moarefi I, Kuriyan J. Crystal structure of the Src family tyrosine kinase Hck. Nature. 1997;385(6617):602-9.\u003c/li\u003e\n\u003cli\u003eBoggon TJ, Eck MJ. Structure and regulation of Src family kinases. Oncogene. 2004;23(48):7918-27.\u003c/li\u003e\n\u003cli\u003eKoboldt DC, Fulton RS, McLellan MD, Schmidt H, Kalicki-Veizer J, McMichael JF, et al. Comprehensive molecular portraits of human breast tumours. Nature. 2012;490(7418):61-70.\u003c/li\u003e\n\u003cli\u003eZhang S, Yu D. Targeting Src family kinases in anti-cancer therapies: turning promise into triumph. Trends in Pharmacological Sciences. 2012;33(3):122-8.\u003c/li\u003e\n\u003cli\u003eWang Y, Ye F, Liang Y, Yang Q. Breast cancer brain metastasis: insight into molecular mechanisms and therapeutic strategies. British Journal of Cancer. 2021;125(8):1056-67.\u003c/li\u003e\n\u003cli\u003eArvanitis CD, Ferraro GB, Jain RK. The blood-brain barrier and blood-tumour barrier in brain tumours and metastases. Nat Rev Cancer. 2020;20(1):26-41.\u003c/li\u003e\n\u003cli\u003ede Bruijn I, Kundra R, Mastrogiacomo B, Tran TN, Sikina L, Mazor T, et al. Analysis and Visualization of Longitudinal Genomic and Clinical Data from the AACR Project GENIE Biopharma Collaborative in cBioPortal. Cancer Res. 2023;83(23):3861-7.\u003c/li\u003e\n\u003cli\u003eGao J, Aksoy BA, Dogrusoz U, Dresdner G, Gross B, Sumer SO, et al. Integrative analysis of complex cancer genomics and clinical profiles using the cBioPortal. Sci Signal. 2013;6(269):pl1.\u003c/li\u003e\n\u003cli\u003eCerami E, Gao J, Dogrusoz U, Gross BE, Sumer SO, Aksoy BA, et al. The cBio cancer genomics portal: an open platform for exploring multidimensional cancer genomics data. Cancer Discov. 2012;2(5):401-4.\u003c/li\u003e\n\u003cli\u003eSuehnholz SP, Nissan MH, Zhang H, Kundra R, Nandakumar S, Lu C, et al. Quantifying the Expanding Landscape of Clinical Actionability for Patients with Cancer. Cancer Discovery. 2024;14(1):49-65.\u003c/li\u003e\n\u003cli\u003eChakravarty D, Gao J, Phillips S, Kundra R, Zhang H, Wang J, et al. OncoKB: A Precision Oncology Knowledge Base. JCO Precision Oncology. 2017(1):1-16.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"bmc-cancer","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bcan","sideBox":"Learn more about [BMC Cancer](http://bmccancer.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/bcan/default.aspx","title":"BMC Cancer","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"breast cancer, cBioPortal, CNS metastasis, domain mapping, Fisher’s exact, Firth logistic, LYN, SH4/Unique, Src-family kinase","lastPublishedDoi":"10.21203/rs.3.rs-7597855/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7597855/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eObjectives: Central nervous system (CNS) metastasis is a major driver of morbidity in metastatic breast cancer, yet the molecular determinants of CNS tropism remain incompletely defined. \u003cem\u003eLYN\u003c/em\u003e, a Src-family kinase integrating receptor tyrosine kinase and integrin signaling, is a biologically plausible mediator of metastatic traits.\u003c/p\u003e\n\u003cp\u003eDesign: We performed a retrospective, multi-study analysis of publicly available breast cancer cohorts aggregated in cBioPortal. After harmonization and de-duplication, \u003cem\u003eLYN\u003c/em\u003e status was determinable in 5,947 invasive carcinoma of no special type (NST) tumors across 29 studies. The primary endpoint was CNS metastasis at any time (Yes/No), harmonized via a prespecified controlled vocabulary (case-insensitive substring mapping). Somatic \u003cem\u003eLYN\u003c/em\u003e variants (coding SNVs/indels) were collapsed to patient-level classes (missense-only; truncating if any nonsense/frameshift/splice). Variants with resolvable positions were mapped to Src-family modules (SH4/Unique, SH3, SH2, SH2–kinase linker, kinase). Two-group comparisons used two-sided Fisher’s exact tests with exact 95% CIs; domain screens used omnibus χ² and Benjamini–Hochberg FDR control. A prespecified Firth logistic model evaluated truncating vs missense within \u003cem\u003eLYN\u003c/em\u003e-mutant tumors.\u003c/p\u003e\n\u003cp\u003eSetting: Public cancer genomics repositories (cBioPortal); multi-institutional cohorts.\u003c/p\u003e\n\u003cp\u003eParticipants: 5,947 tumors across multi-study cohorts with \u003cem\u003eLYN\u003c/em\u003e status available.\u003c/p\u003e\n\u003cp\u003eInterventions: None.\u003c/p\u003e\n\u003cp\u003eMain outcome measures: Primary: ever-CNS metastasis (yes/no). Secondary: distribution of \u003cem\u003eLYN\u003c/em\u003evariant classes and domains (SH4/Unique, SH3, SH2, linker, kinase).\u003c/p\u003e\n\u003cp\u003eResults: CNS metastasis occurred in 5/46 (10.9%) \u003cem\u003eLYN\u003c/em\u003e-mutated tumors vs 110/5,901 (1.9%) \u003cem\u003eLYN\u003c/em\u003e wild-type tumors (OR = 6.42; 95% CI, 2.49–16.56; p = 0.0018). Within \u003cem\u003eLYN\u003c/em\u003e-mutant cases, domain distributions differed by CNS status (omnibus χ² p ≈ 0.014); a one-versus-rest signal at the SH4/Unique N-terminus was nominally significant and borderline after FDR (unadjusted p ≈ 0.010; q ≈ 0.052; small in-domain n = 3). By mutation class, truncating vs missense showed a higher CNS-positive proportion (28.6% vs 7.9%) but did not reach significance (Fisher p = 0.166; alternatively framed OR = 4.22; exact 95% CI, 0.58–30.75; p = 0.182). Firth estimates were directionally consistent with wide profile CIs under sparse counts.\u003c/p\u003e\n\u003cp\u003eConclusions: Across pooled cohorts, \u003cem\u003eLYN\u003c/em\u003e mutation is associated with increased odds of CNS metastasis, and domain context appears informative, with a small-sample, FDR-borderline enrichment at the SH4/Unique N-terminus. The truncating-class signal is exploratory given limited power. These data prioritize domain-aware \u003cem\u003eLYN\u003c/em\u003eannotation for independent validation and mechanistic follow-up focused on membrane targeting, endothelial adhesion, and blood–brain barrier traversal.\u003c/p\u003e\n\u003cp\u003eTrial registration: Not applicable.\u003c/p\u003e","manuscriptTitle":"LYN Mutations in Breast Cancer: Association with Central Nervous System Metastasis and Domain-Level Insights","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-10-20 09:42:33","doi":"10.21203/rs.3.rs-7597855/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-10-31T11:07:24+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-10-21T07:52:09+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-10-17T14:26:07+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"150182871422247376108884417020335801196","date":"2025-10-10T12:43:12+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"229367627557595098582930056386950060088","date":"2025-10-08T13:43:29+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-10-08T10:23:55+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"195159781274447841934156486306629020679","date":"2025-10-07T11:54:20+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-10-07T10:27:28+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-10-07T05:50:57+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2025-09-18T08:25:43+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-09-17T08:07:56+00:00","index":"","fulltext":""},{"type":"submitted","content":"BMC Cancer","date":"2025-09-17T08:03:54+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"bmc-cancer","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bcan","sideBox":"Learn more about [BMC Cancer](http://bmccancer.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/bcan/default.aspx","title":"BMC Cancer","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"40c49573-dd38-41a4-a36a-011eb75b0ca6","owner":[],"postedDate":"October 20th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2025-12-08T16:10:59+00:00","versionOfRecord":{"articleIdentity":"rs-7597855","link":"https://doi.org/10.1186/s12885-025-15398-w","journal":{"identity":"bmc-cancer","isVorOnly":false,"title":"BMC Cancer"},"publishedOn":"2025-12-05 15:57:42","publishedOnDateReadable":"December 5th, 2025"},"versionCreatedAt":"2025-10-20 09:42:33","video":"","vorDoi":"10.1186/s12885-025-15398-w","vorDoiUrl":"https://doi.org/10.1186/s12885-025-15398-w","workflowStages":[]},"version":"v1","identity":"rs-7597855","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7597855","identity":"rs-7597855","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}