Cancer Phenotypes of Hypomorphic versus Non-Functional TP53 Variants in Li-Fraumeni Syndrome
Annual Clinical Genetics Meeting, March 2026
Alison S. Levine, Dana-Farber Cancer Institute, Division of Cancer Genetics and Prevention, Alison S. Levine, Adela Rodriguez-Hernandez, MD, Miki Horiguchi, PhD, Isabel Anez, MS, CGC, Sophie Cahill, BS, Judy E. Garber, MD, MPH, Huma Q. Rana, MD, FACMG
Introduction: Germline TP53 pathogenic variants (PVs) are associated with Li-Fraumeni syndrome (LFS). Recognition of TP53 variants that lead to attenuated LFS, also referred to as hypomorphic variants, has increased in recent years. These variants frequently suffer from discordant classification, leading to confusion and inconsistent or unclear management recommendations for families. Hypomorphic variants are increasingly identified through broad germline testing in individuals with cancer. They have been characterized using functional assays or reported phenotypes, but no systematic approach has leveraged detailed clinical pedigrees. Understanding how hypomorphic variants differ clinically from classic dominant-negative or loss-of-function (DNE/LOF) TP53 variants is essential for accurate risk assessment.
Methods: This retrospective cohort study included 431 individuals evaluated through a single center cancer genetics clinic and tested between 1998 and 2025. The cohort was identified using a genotype-first approach. Hypomorphic TP53 variants were defined as those demonstrating partial transcriptional activity, as classified by Kato et al. A query of the NCI TP53 database (2024) identified 1,005 variants, 35 of which were present in the DFCI patient database. Variants observed in ≥4 unrelated families with clinical data were selected for analysis, yielding 6 variants: p.Arg181His, p.Arg337His, p.Arg156His, p.Arg181Cys, p.Arg282Gln, and p.Arg156Cys. The p.Gly334Arg variant was also included given its established hypomorphic function. Clinical characteristics were summarized using descriptive statistics. Bivariate associations between variant type (hypomorphic vs. DNE/LOF) and clinical features were evaluated using permutation tests to account for within-family correlation. For each variant category, cumulative proportions of individuals diagnosed with cancer by ages 20, 40, and 60 years were estimated, and differences between categories were assessed using permutation tests accounting for family clustering.
Results: A total of 389 individuals with TP53 variants were included in the final analytic cohort. Median age at testing was 42 years, and 65.6% were female. Overall, 217 individuals had a cancer diagnosis, occurring in 47.8% of hypomorphic carriers compared with 74.8% of those with DNE/LOF variants (p<0.001). LFS phenotype differed markedly by variant type. No individuals with hypomorphic variants met classic LFS criteria, compared with 23.5% in the DNE/LOF group (p<0.001). Chompret criteria were met in 27.4% vs 65.2% respectively (p=0.001). Attenuated LFS categories (attenuated LFS, attenuated LFS carrier, incidental LFS per Kratz) were present in 63.8% of hypomorphic carriers versus 33.9% of DNE/LOF carriers. Median age at first cancer diagnosis was older in hypomorphic carriers (48 vs 38 years; p<0.001). The age at first cancer diagnosis differed significantly between the hypomorphic group and DNE/LOF group respectively with the following proportions: 8.4% vs 19.8% by age 20, 30.5% vs 60.5% by age 40, and 77.9% vs 95.3% by age 60. Hypomorphic carriers had fewer multiple primary cancers (16.1% vs 40.0%; p<0.001) and fewer LFS core cancers (2.2% vs 6.1%; p=0.036). They also showed lower frequencies of central nervous system tumors (1.8% vs 8.7%; p=0.002), colorectal (1.1% vs 10.4%; p<0.001), gastric (0% vs 1.7%; p=0.005), lung (2.9% vs 10.4%; p=0.002), ovarian (1.5% vs 5.2%; p=0.032), and urothelial cancer (0.4% vs 3.5%; p=0.005). For breast cancer, age at first diagnosis was older in hypomorphic carriers (50 vs 39 years; p<0.001), with no subtype differences. Risk-reducing surgeries did not significantly differ by variant type. Among evaluable females, 7.2% had risk-reducing mastectomy, 25.8% contralateral prophylactic mastectomy, and 19.8% salpingo-oophorectomy.
Conclusion: In this large genotype-first LFS cohort, hypomorphic TP53 variants exhibited a clinical phenotype distinct from that of non-functional variants, with lower penetrance, older age at diagnosis, and differences in tumor spectrum. Incorporating functional class and family-level phenotypes into clinical interpretation may lead to more accurate counseling and personalized management for affected families.
