Identification of genetic variants in the FGB gene associated with congenital hypofibrinogenemia with divergent clinical phenotype Authors: Kristina Maria Belakova, Rosanna Asselta, Sonia Caccia, Zuzana Kolkova, Dusan Loderer, Miroslava Drotarova, Monika Brunclikova, Sohaib Mukhtar Agouba, Veronika Voskova Gemelova, Jan Stasko, Jan Mikler, Tomas Simurda Published: 2026 ABSTRACT Congenital fibrinogen disorders are rare and clinically heterogeneous conditions that may manifest with both bleeding and thrombotic complications. The relationship between genotype and phenotype remains incompletely understood, particularly in cases involving co-inherited variants with potentially opposing functional effects. We investigated two families with hypofibrinogenemia using coagulation assays, rotational thromboelastometry (ROTEM), genetic analysis, and protein modelling. In the first family, an 18-year-old patient with mild bleeding carried two heterozygous FGB variants: c.794 C > T (p.Pro265Leu) and c.1102T > C (p.Tyr368His), and showed reduced clot firmness on ROTEM, consistent with a quantitative fibrinogen defect. Segregation analysis indicated that p.Pro265Leu was associated with decreased fibrinogen levels, whereas p.Tyr368His was present in an asymptomatic relative. In the second family, an 11-year-old patient and her father carrying p.Tyr368His presented with asymptomatic hypofibrinogenemia and relatively preserved ROTEM parameters, suggesting a limited impact of this variant on fibrin-based clot formation. Protein modelling indicated that p.Tyr368His had the more pronounced predicted structural effect, showing moderate local destabilization in both isolated and assembled models, whereas p.Pro265Leu showed a mild and context-dependent effect that was largely compensated in the trimeric fibrinogen structure. These findings demonstrate that co-inherited fibrinogen variants may exert distinct and potentially opposing effects on clot formation, contributing to variable clinical phenotypes. Integration of genetic data with global functional assays such as ROTEM and with structural modelling provides important insights into variant-specific effects and may improve the interpretation of pathogenicity and support individualized clinical management. Introduction Fibrinogen is a key plasma glycoprotein essential for hemostasis, serving as the precursor of fibrin and contributing to clot formation and stability [1, 2]. Congenital fibrinogen disorders (CFDs) are rare inherited conditions caused by pathogenic variants in the fibrinogen genes (FGA, FGB, FGG), affecting fibrinogen synthesis, secretion, or function [1, 3]. CFDs are classified into quantitative defects (afibrinogenemia and hypofibrinogenemia) and qualitative defects (dysfibrinogenemia and hypodysfibrinogenemia) [4]. Clinical manifestations are highly variable, ranging from asymptomatic cases to bleeding or thrombotic complications [3, 5, 6]. Notably, thrombotic events may occur even in patients with reduced fibrinogen levels, highlighting the complex relationship between fibrinogen quantity, structure, and function [5, 6]. Establishing genotype–phenotype correlations in CFDs remains challenging, as similar variants may result in different clinical presentations [1, 7]. In particular, missense variants in the FGB gene may differentially affect fibrinogen secretion or alter fibrin structure, contributing to heterogeneous phenotypes [3]. However, the clinical impact of co-inherited fibrinogen variants with potentially opposing functional effects remains poorly understood. In this study, we describe two families with hypofibrinogenemia carrying FGB variants and integrate clinical, laboratory, and structural data, including protein modelling, to explore the mechanisms underlying their variable phenotypic expression. Materials and methods Our study was conducted in accordance with the Declaration of Helsinki. It was also approved by the institutional Ethics committee of the Comenius University in Bratislava, Jessenius Faculty of Medicine, and informed consent was obtained. Coagulation screening tests Venous blood was collected into tubes containing 0.125 M trisodium citrate at a 1:9 ratio (citrate: blood, pH 7.3). Plasma was obtained by blood centrifugation at 1500×g for 10 min. Prothrombin time (PT), activated partial thromboplastin time (APTT), thrombin time (TT), and reptilase time (RT) (all reagents from IL-Instrumentation Laboratory; Bedford, Massachusetts, USA) were measured using the ACL TOP 550 CTS automatic analyser (Beckman-Coulter, Brea, California, USA). Plasma fibrinogen concentration was measured by the Clauss method (IL-Instrumentation Laboratory) and fibrinogen antigen levels were determined by a turbidimetric latex immunoassay (LIA) (Hyphen BioMed, West Chester, Ohio, USA) on plasma samples. Rotational thromboelastometry Rotational thromboelastometry (ROTEM) was performed using a ROTEM® delta (Tem Innovations GmbH, Munich, Germany) from citrated samples of blood. All samples were processed within a maximum of 30 min after withdrawal. Three routine ROTEM assays: extrinsically activated thromboelastometric test (EXTEM), intrinsically activated thromboelastometric test (INTEM), and extrinsically activated thromboelastometric test (FIBTEM) were conducted to assess clotting time (CT), clot formation time (CFT), amplitude representing the clot firmness in 10 min after CT (A10) or 20 min after CT (A20), and maximum clot firmness (MCF) according to the manufacturer’s instructions [8, 9]. Genetic analysis DNA was isolated from 2 ml of whole blood by using a QIAamp DNA Blood Midi Kit (Qiagen, Hilden, Germany). Genetic analysis of fibrinogen FGB, FGA, and FGG genes was performed by Next-Generation Sequencing (NGS) on 10 ng of genomic DNA using the AmpliSeq Custom DNA Panel for Illumina (Illumina, San Diego, California, USA). Libraries were prepared using the AmpliSeq Library PLUS for Illumina, following the manufacturer’s instructions. The final pool of libraries was denatured and sequenced on the MiSeq sequencer (2 × 151 bp; Illumina). Fastq mapping to the reference genome (hg38) and germline variant calling were performed using the Identify DNA Germline Variants workflow in CLC Genomics Workbench v.20 (Qiagen). The minimum read depth required for variant analysis was 250x. Filtered VCF files were uploaded and annotated using the Ensembl Variant Effect Predictor (VEP; https://www.ensembl.org/info/docs/tools/vep/index.html). The identified variants were confirmed by PCR amplification of the relevant exon(s) followed by direct Sanger sequencing using the BigDye Terminator v1.1 Cycle Sequencing Kit and an ABI 3500 Genetic Analyzer (Applied Biosystems, Waltham, MA, USA). Protein modelling Protein modelling was performed by means of FoldX [10]. Structural analyses were carried out using both the isolated fibrinogen Bβ chain and the Aα/Bβ/γ-chain assembly in order to assess the effects of the identified amino acid substitutions on local folding stability and interchain interactions, respectively. For each variant, the corresponding mutant models were generated from the crystal structure 1FZA.pdb, after structure repair, and the predicted energetic effect was expressed as ΔΔG relative to the wild-type model Molecular graphics and analyses were performed with UCSF Chimera, developed by the Resource for Biocomputing, Visualization, and Informatics at the University of California, San Francisco, with support from NIH P41-GM103311 [11]. Results Family 1 An 18-year-old male was diagnosed with hypofibrinogenemia during preoperative evaluation at the age of 6 years. Laboratory testing revealed reduced fibrinogen activity (0.9 g/L; reference range 1.8–4.2 g/L) and antigen levels (1.2 g/L; reference range 1.8–4.2 g/L), while PT, APTT, TT, and RT were within normal limits (Table 1). Clinically, the patient exhibited a mild bleeding phenotype, including perioperative bleeding following adenoidectomy and occasional mucocutaneous bleeding episodes. ROTEM analysis demonstrated reduced clot firmness, with decreased FIBTEM A10 (3 mm; reference range 7–23 mm) and MCF (3 mm; reference range 9–25 mm), as well as reduced EXTEM A10 (33 mm; reference range 43–65 mm) and MCF (43 mm; reference range 50–72 mm) (Fig. 1A).Notably, mildly decreased clot firmness parameters were also observed in the patient’s father, with EXTEM A10 of 41 mm and FIBTEM MCF of 7 mm. ROTEM analysis was not performed in the patient’s mother. Fig. 1Results of rotational thromboelastometry (EXTEM, INTEM, FIBTEM) in patient 1 (A) and patient 2 (B). Abbreviation: A10 and A20, clot amplitude 10 and 20 min after CT; CFT, clot formation time; CT, clotting time; MCF, maximum clot firmness Results of rotational thromboelastometry (EXTEM, INTEM, FIBTEM) in patient 1 (A) and patient 2 (B). Abbreviation: A10 and A20, clot amplitude 10 and 20 min after CT; CFT, clot formation time; CT, clotting time; MCF, maximum clot firmness Genetic analysis identified two heterozygous variants in the FGB gene: c.794 C > T (p.Pro265Leu) and c.1102T > C (p.Tyr368His), the latter corresponding to the variant Fibrinogen Martin II associated with a thrombotic phenotype [3]. Segregation analysis demonstrated that p.Pro265Leu was present in the mother and was associated with reduced fibrinogen levels. The p.Tyr368His variant was identified in the father, who exhibited borderline fibrinogen levels. In the brother, the mutation was not detected and laboratory parameters were within the normal range (Fig. 2A). Fig. 2Pedigree chart of patient 1 (A) and patient 2 (B) with their relatives Pedigree chart of patient 1 (A) and patient 2 (B) with their relatives The fibrinogen activity-to-antigen ratio (Clauss/LIA) in Patient 1 was 0.75, consistent with a quantitative fibrinogen defect. Family 2 An 11-year-old female was diagnosed with mild hypofibrinogenemia during preoperative evaluation. Laboratory testing showed reduced fibrinogen activity (1.22 g/L) and antigen levels (1.2 g/L), with otherwise normal coagulation parameters except for mildly prolonged TT (19.9 s) (Table 1). Despite reduced fibrinogen levels, the patient had no history of bleeding complications, including during surgical intervention. ROTEM analysis revealed mildly reduced clot firmness, with FIBTEM A10 of 6 mm and MCF of 6 mm, while EXTEM parameters were only slightly decreased (Fig. 1B). Genetic analysis identified a heterozygous FGB variant c.1102T > C (p.Tyr368His), corresponding to Fibrinogen Martin II [3]. The same variant was detected in the patient’s father, who also exhibited reduced fibrinogen levels but normal ROTEM parameters. The fibrinogen activity-to-antigen ratio in Patient 2 was 1.02, indicating preserved functional activity relative to antigen levels (Fig. 2B). Table 1Coagulation tests of family 1 and family 2Family 1PT (s)APTT (s)TT (s)Fbg Clauss (g/L)Fbg LIA (g/L)PLT(×10⁹/L) Patient 112.430.914.20.901.2246 Father11.028.014.81.812.0228 Mother11.330.214.21.611.5193 Brother11.427.215.12.342.1255Family 2 PT (s) APTT (s) TT (s) Fbg Clauss (g/L) Fbg LIA (g/L) PLT(×10⁹/L) Patient 211.530.819.91.221.2242 Father11.327.515.41.541.2353 Mother10.229.912.33.50-243 Brother11.236.916.52.53-357Normal range10.4–12.622.0–32.012.0–18.01.80–4.201.8–4.2140–400PT prothrombin time, aPTT activated partial thromboplastin time, Fbg fibrinogen, TT thrombin time, RT reptilase time, LIA latex immunoassay, PLT platelets Table 1Coagulation tests of family 1 and family 2Family 1PT (s)APTT (s)TT (s)Fbg Clauss (g/L)Fbg LIA (g/L)PLT(×10⁹/L) Patient 112.430.914.20.901.2246 Father11.028.014.81.812.0228 Mother11.330.214.21.611.5193 Brother11.427.215.12.342.1255Family 2 PT (s) APTT (s) TT (s) Fbg Clauss (g/L) Fbg LIA (g/L) PLT(×10⁹/L) Patient 211.530.819.91.221.2242 Father11.327.515.41.541.2353 Mother10.229.912.33.50-243 Brother11.236.916.52.53-357Normal range10.4–12.622.0–32.012.0–18.01.80–4.201.8–4.2140–400PT prothrombin time, aPTT activated partial thromboplastin time, Fbg fibrinogen, TT thrombin time, RT reptilase time, LIA latex immunoassay, PLT platelets Coagulation tests of family 1 and family 2 PLT (×10⁹/L) PT prothrombin time, aPTT activated partial thromboplastin time, Fbg fibrinogen, TT thrombin time, RT reptilase time, LIA latex immunoassay, PLT platelets Protein modelling To investigate the structural consequences of the identified FGB variants, protein modelling was performed using FoldX on both the isolated fibrinogen Bβ-chain module and on the trimeric structure comprising the Aα, Bβ, and γ chains. These analyses were used to estimate the effects of the amino acid substitutions on local folding stability and on interchain interactions within the assembled fibrinogen molecule (Fig. 3). Fig. 3(a) Localization of p.Pro265 and p.Tyr368 (red spheres) in the b -nodule (yellow) within the fibrinogen structure (b1) close-up view of the salt bridges network centered around residue Tyr368 (wt) and (b2) His368 (mutant) (c) close-up view of position 265 at the interface of b and g -nodules:wild-type Pro (yellow) and mutant Leu (red) are superimposed. (a) Localization of p.Pro265 and p.Tyr368 (red spheres) in the b -nodule (yellow) within the fibrinogen structure (b1) close-up view of the salt bridges network centered around residue Tyr368 (wt) and (b2) His368 (mutant) (c) close-up view of position 265 at the interface of b and g -nodules:wild-type Pro (yellow) and mutant Leu (red) are superimposed. For the p.Tyr368His variant (Bβ Y338H in the mature chain), FoldX analysis consistently predicted a destabilizing effect. Comparable results were obtained when modelling the whole fibrinogen structure or the isolated Bβ chain, indicating that the structural perturbation is predominantly local. In the Bβ-chain model, the predicted ΔΔG values across four independent runs have a mean of + 2.14 ± 0.06 kcal/mol, consistent with a moderate destabilizing effect. Structural comparison between wild-type and mutant models showed that the side-chain environment was altered. In the wild-type structure, Tyr338 formed a close contact with Asp297, compatible with a stabilizing hydrogen bond, while in the mutant model this interaction was lost and the side chain of Asp297 adopted a different orientation. The p.Pro265Leu variant (Bβ P235L in the mature chain) showed a different behaviour depending on the structural context used for modelling. In the isolated Bβ-chain model, the variant was predicted to have a mild destabilizing effect, with a mean ΔΔG value of + 0.94 ± 0.09 kcal/mol. In contrast, when the same substitution was analysed in the context of the trimeric assembly, the predicted energetic effect became essentially neutral. The calculated ΔΔG mean value decreased to + 0.054 ± 0.142 kcal/mol. Discussion Congenital hypofibrinogenemia is associated with marked clinical heterogeneity, ranging from asymptomatic presentation to bleeding or thrombotic complications [5, 6]. This variability reflects the complex relationship between fibrinogen quantity, structure, and function and contributes to the difficulty of establishing clear genotype–phenotype correlations [7, 12–14]. Pathogenic variants in fibrinogen genes may differentially affect fibrinogen synthesis, secretion, or fibrin structure, contributing to variable laboratory and clinical phenotypes [3, 8]. In particular, FGB variants are frequently associated with quantitative fibrinogen defects due to impaired fibrinogen assembly or secretion. Accurate diagnosis and additional analyses in research laboratories working on fibrinogen and fibrin clot structure can help to better assess the patient phenotype and consequently the appropriate treatment [7]. In addition, the personal and familial history is a major determinant of the patient’s phenotype, as in thromboembolic or other congenital coagulation disorders [15]. Our first patient was diagnosed with moderate hypofibrinogenemia before surgical intervention. His coagulation assays showed reduced fibrinogen activity and antigen level, and ROTEM analyses confirmed decreased clot amplitude and maximum clot firmness. Genetic analysis showed 2 pathogenic variants: a heterozygous single nucleotide variant in the FGB c.794 C > T (p.Pro265Leu) associated with bleeding potential, and a heterozygous thrombophilic Bβ missense variant in the FGB c.1102T > C (p.Tyr368His), also known as Fibrinogen Martin II [3]. This combined genetic background may contribute to the observed laboratory phenotype and perioperative bleeding tendency, consistent with an additive effect on fibrinogen quantity and clot mechanics. The second patient also exhibited decreased fibrinogen activity (1.22 g/L) preoperatively. Although carrying the same Bβ p.Tyr368His mutation, the individual had no prior clinical manifestations of its thrombophilic potential. Together, these cases further illustrate incomplete penetrance and variable expressivity, emphasizing that interpretation of rare fibrinogen variants benefits from integration of clinical history, family segregation, global assays such as ROTEM, and structural modelling. An important aspect of our findings is the functional correlation between genotype and global coagulation assessment by ROTEM. In Patient 1, markedly reduced FIBTEM and EXTEM clot firmness parameters were observed in association with the combined presence of p.Pro265Leu and p.Tyr368His. This ROTEM profile is consistent with impaired fibrin-based clot formation and suggests a predominantly quantitative fibrinogen defect. Although p.Tyr368His has previously been associated with thrombophilic properties, its presence did not appear to offset the reduced clot firmness observed in this patient, indicating that the effect of reduced fibrinogen levels may predominate in this combined genetic context. Protein modelling nevertheless showed that p.Pro265Leu has only a mild and strongly context-dependent structural effect. This context dependency was also reflected in the structural modelling results. While p.Pro265Leu showed a mild destabilizing effect in the isolated β-chain model, its energetic impact became nearly neutral in the assembled fibrinogen structure, suggesting partial compensation through interchain interactions. By contrast, p.Tyr368His consistently demonstrated a stronger destabilizing effect across different modelling conditions, supporting a more pronounced local structural perturbation. In the isolated β-nodule it was weakly destabilizing, mainly through loss of local conformational restraint rather than disruption of a defined side-chain network, whereas in the trimeric fibrinogen model its effect became essentially neutral, with partial compensation at the β-γ-nodules interface. Overall, these findings are consistent with previous reports suggesting that the p.Pro265Leu substitution in the Bβ-chain does not exert a major independent effect on fibrinogen stability and may instead act as a limited or modifier variant rather than as a strong isolated pathogenic determinant [16, 17]. Although p.Pro265Leu is not a novel variant and has previously been associated with fibrinogen disorders, its clinical and functional significance remains incompletely characterized, and the present findings further support its variable phenotypic expression. In contrast to Patient 1, Patient 2 showed only mild reductions in ROTEM clot firmness despite the presence of the p.Tyr368His variant. This finding suggests that the variant alone may not substantially impair fibrin polymerization or clot stability and may instead exert a more limited or modulatory effect on fibrin structure. The relatively preserved ROTEM profile despite reduced fibrinogen levels further illustrates the functional heterogeneity associated with fibrinogen variants. By contrast, protein modelling identified p.Tyr368His as the variant with the clearest predicted structural impact. FoldX consistently indicated a destabilizing effect in both the isolated β-nodule and the assembled fibrinogen model, supporting a predominantly local perturbation. Structural analysis suggested disruption of a Tyr-centered interaction network in the β-nodule, including loss of contact with Asp297 and altered orientation of nearby side chains. These findings are consistent with the high conservation of this residue and with previous evidence linking p.Tyr368His in homozygosity to hypofibrinogenemia with recurrent venous thrombosis [3, 18]. However, the relatively preserved ROTEM profile observed in our heterozygous carrier, despite reduced fibrinogen levels, suggests that this structural perturbation may have a more limited functional expression in heterozygosity than predicted by modelling alone. This interpretation is supported by recent data showing variable clinical expression among heterozygous carriers of p.Tyr368His, including bleeding, thrombosis, and asymptomatic presentation [19], in keeping with the broader phenotypic heterogeneity reported in congenital hypofibrinogenemia [20]. Taken together, these observations suggest that p.Tyr368His is structurally relevant, but that its phenotypic expression may depend on zygosity and on additional modifying factors. These findings highlight that ROTEM may provide a valuable functional readout linking genetic variants with their phenotypic expression, particularly in cases where variants with potentially opposing biological effects coexist. Another limitation is that ROTEM analyses were based on single time-point measurements obtained during routine clinical assessment, and repeated analyses were not available to assess intra-individual reproducibility. A limitation of this study is the lack of direct functional assays evaluating fibrin clot structure and fibrin polymerization. Structural analyses were limited to in silico modelling of fibrinogen, and the potential effects of the identified variants on fibrin architecture were therefore inferred indirectly from ROTEM findings and computational modelling. In addition, the combined structural effect of the co-inherited variants was not specifically modelled as a single integrated fibrinogen/fibrin complex. Taken together, our findings indicate that co-inherited fibrinogen variants may exert counterbalancing effects, resulting in phenotypes that cannot be predicted from individual variants alone. The contrasting ROTEM profiles observed in our patients emphasize the value of integrating functional and genetic data, which may be essential for accurate clinical interpretation and individualized risk assessment in congenital fibrinogen disorders. These findings underscore the limitation of predicting clinical phenotype based solely on genotype, highlighting the need for integrated functional and molecular assessment. Given the limited number of investigated cases, the findings should be interpreted as exploratory and hypothesis-generating. Conclusions This study identified a combination of pathogenic variants in patients with congenital hypofibrinogenemia affecting the clinical phenotype. These case reports underscore the complexity of the hemostatic system and the importance of considering individual patient characteristics beyond standard laboratory tests. The identification of the precise genetic defect of congenital fibrinogen disorders can help to test other at-risk individuals, to understand the correlation between genotype and phenotype, to assist in therapeutic choices, and for the development of new specific treatments, such as gene therapy. Our study provides valuable insights for correlating functional and structural changes in mutant fibrinogen with posttranslational processing and biomechanical properties.