2026 ASH®
Annual Meeting & Exposition December 12-15, 2026 ▪ New Orleans, Louisiana

Paper Details

Abstracts, authors, affiliations, and disclosures.

Sunday, December 13, 2026
12:00 p.m. - 01:30 p.m.
MCCNO - 217-219
Treatment Considerations, Biological Therapies, Antibody Therapy, Bispecific Antibody Therapy, Infusion, Monoclonal Antibody Therapy, Non-Biological Therapies, Coagulant Drugs, Women’s Health, Biological Processes, Genomics, Diseases, Bleeding and Clotting, Hemophilia, Research, Fundamental Science, Translational Research, Clinical Research, Clinical Trials, Registries, Study Population, Human, Pediatric
PAPER

601 Hemophilia A genotype status predicts bleeding in females, while FVIII levels fail to identify a “safe” female FVIII threshold in my life, our future.

Hui Xiao Chao, MD, PhD, Tom W Walsh2, Kerry Lannert, MT(ASCP)3, Danielle N Drury-Stewart, PhD4, G Shellye Horowitz, MA5, Shelley Nakaya Fletcher6, Jill Johnsen, MD7
1 University of Washington, Seattle, Washington, United States
2 University of Washington, Seattle, Washington, United States
3 University of Washington, Seattle, Washington, United States
4 University of Washington, Seattle, Washington, United States
5 University of Washington, Seattle, Washington, United States
6 Bloodworks Northwest Research Institute, Seattle, Washington, United States
7 University of Washington, Seattle, Washington, United States

Introduction: Hemophilia A is an X-linked inherited bleeding disorder caused by low coagulation factor VIII (FVIII). FVIII activity levels <40% define hemophilia A. Unlike males, mounting evidence suggests that females with FVIII ≥40% may still experience excessive bleeding. We sought to test the hypotheses that F8 genotype is associated with excessive bleeding in females, and that there is a FVIII threshold above which there is protection against bleeding in females.

Methods: Data and samples were obtained with approval from the My Life, Our Future (MLOF) Research Repository. MLOF was a U.S. hemophilia genotyping program where participants could consent to research. Data were obtained from the Bloodworks genotyping laboratory and American Thrombosis and Hemostasis Network (ATHN), including genotype, age, sex, reported baseline FVIII activity, ISTH-SSC Bleeding Assessment Tool (BAT) score, and treatment history. Groups were compared using Wilcoxon rank-sum or Fisher’s exact tests.

Results: We identified 808 females with genotype, age, FVIII levels, and ISTH-BAT scores. 645 (79.8%) had a hemophilia A-causing genotype. FVIII levels were lower in genotype-positive compared with genotype-negative females (mean 71.7% vs 122.1%, p<0.0001), as expected. 18.0% of genotype-positive females had FVIII <40%, compared to 5.5% without a genotype.

Total ISTH-BAT scores were higher in genotype-positive females than genotype-negative females (3.36 vs 2.40, p=0.00058). Age-adjusted abnormal BAT scores were more common in genotype-positive than genotype-negative females (25.0% vs 17.2%, p=0.038). Surgical bleeding was most common (36.7%), followed by epistaxis (27.8%), minor wound bleeding (19.2%), and hematuria (19.1%).

FVIII activity did not correlate with BAT score. The highest bleeding scores were associated with FVIII <10%. At FVIII >10%, the abnormal bleeding rate flattened, remaining at >15% even with FVIII ≥100%. For individuals with FVIII between 40% and 100%, there was no correlation between FVIII level and BAT score (Spearman’s ρ=−0.05, p=0.327). Applying a FVIII threshold of <40% demonstrated poor sensitivity (25.5%) in identifying individuals with a high BAT score. This threshold missed 120 of 529 (22.7%) females with FVIII ≥40% who had high BAT scores. For comparison, 35.3% of females with FVIII <40% had BAT high scores.

To evaluate for clinically significant bleeding, treatment history was analyzed for the 572 females with treatment history and FVIII levels available. In genotype-positive females with FVIII >40%, 210/299 (70.2%) received hemophilia-specific treatments. Notably, the genotype-negative group had a high proportion of individuals (31.4%) receiving von Willebrand factor replacement, suggesting a diagnosis of VWD, compared to 7.1% in the genotype-positive group (p<0.0001).

We analyzed F8 genotypes for association with bleeding. Females with synonymous variants were less likely to have a high BAT score (6.5%) compared with missense (25.6%, p=0.014) and all null variants, including nonsense (34.9%, p=0.0047), frameshift (31.8%, p=0.0053), and intron 22 inversions (25.0%, p=0.019). Missense and null variants had similar bleeding profiles across all 14 BAT domains. A logistic regression model among females with FVIII ≥ 40% found that all the non-synonymous variant classes analyzed had >10-fold increased odds of abnormal bleeding compared to the general population.

The most common genotypes were intron 22 inversion (n=164) and c.1569G>T (p.Leu523Leu) (n=31). Females with the inversion had significantly higher BAT scores (3.30 vs 1.45, p=0.022) and were more likely to have an abnormal BAT (25.0% vs 6.45%, p=0.019), despite similar FVIII levels (72.3% vs 72.1%, p=0.76).

Conclusion: Excessive bleeding is common in females with a hemophilia-causing genotype. Bleeding remained common despite FVIII ≥ 40% (22.7% vs. 35.3% for FVIII <40%). The 40% FVIII threshold demonstrates poor sensitivity (25.5%) to identify females with abnormal bleeding. The presence of a genotype is strongly associated with bleeding across the range of FVIII levels. There is some evidence that DNA variant class and individual F8 genotypes were associated with bleeding phenotype independent of FVIII level. Our findings suggest incorporating genotype information into bleeding risk assessment for females at risk in hemophilia A. Disappointingly, we could not identify an FVIII level threshold above which there is protection against bleeding in females.

Hui Xiao Chao, MD, PhD: No relevant conflicts of interest to declare.