Table S7: Differential expression analysis of 328 p53 target genes in HCT116 p53 +/-, R181H/-. R181C/- cells after 24 hours of Nutlin-3a compared to vehicle.

Figure S10. R181-mutant p53 goes to the mitochondria and induces cytochrome c upon genotoxic stress.

Figure S3. R181H tumors undergo loss of heterozygosity.

Genetic testing is increasingly recommended for adolescents and young adults (AYAs) with cancer; however, no AYA-specific models for cancer risk communication have been developed. We developed a chatbot-based patient- and family-centered cancer risk communication tool, the AYA-RISE (AYA-Risk Information and Screening Education) intervention for AYAs aged 12–24 years. The intervention was developed together with AYAs with cancer risk syndromes, their family members, and clinicians, using Invitae’s Gia® chatbot (study Phase 1). 17 AYAs participated in a group discussion and completed surveys for input regarding development. Content was developed for 9 different syndromes. AYA-RISE was then refined after iterative input (Phase 2) and evaluated for feasibility and preliminary outcomes in a pilot study (Phase 3). 100

Table S4: History of cancer and age of onset in carriers of R181, DN/LOF, and hypomorphic variants in the patient cohort.

Table S2A. Antibodies, cell lines, and drugs used in this study Table S2B: Software and algorithms used in this study.

Figure S2. Representative Pedigrees of families with the TP53 p.R181H pathogenic germline variant from Penn Medicine (Penn), Dana-Farber Cancer Institute (DFCI), and Huntsman Cancer Institute (HCI). Probands are indicated by the black arrowhead; slashes indicated deceased status.

Table S6: Differential expression analysis of 328 p53 target genes in HCT116 p53 +/-, R181H/-. R181C/- cells after 8 hours of Nutlin-3a compared to vehicle.

Table S1. Detailed clinical information on R181H/C probands from three academic medical centers.

Table S11: Differential expression analysis of 328 p53 target genes in MCF7 p53 +/-, R181H/-. R181C/- cells after 24 hours of Nutlin-3a compared to vehicle.

e13571 Background: Mainstream germline testing, where oncology providers obtain consent and order testing directly, facilitates timely access to genetic testing for treatment decisions. The Veterans Health Administration promoted mainstreaming of a 62-gene multi-cancer germline testing panel for certain cancer diagnoses: breast, high-grade or metastatic prostate, ovarian/fallopian tube/primary peritoneal/serous uterine, pancreatic/ampullary adenocarcinoma, medullary thyroid, pheochromocytoma/paraganglioma, colorectal cancer diagnosed before age 50, and mesothelioma. We retrospectively evaluated conformance to guidance for adoption of mainstream testing. Methods: We analyzed all germline genetic tests ordered across VA facilities from February 2023 through December 2025. Test orders were categorized as conforming oncology (mainstream-eligible cancer diagnosis), non-conforming oncology (other cancer diagnosis), or non-oncology tests. For orders placed by oncology providers, clinical appropriateness was determined by ICD-10 code alignment with specified mainstream-eligible cancer types. Conformance to guidance was defined as ordering the 62-gene panel for patients with one of the specified mainstream-eligible cancer diagnoses. Primary outcome was rate of conformance to guidance in our mainstream testing model. Results: Among 13,339 genetic tests ordered by 204 providers across 143 facilities, 3,306 (24.7%) used mainstream testing and 10,033 (75.3%) used traditional genetics. Of mainstream orders, 3,130 (94.7%) were guidance conforming and placed by 109 providers, 161 (4.9%) were non-conforming oncology tests placed by 34 providers, and 15 (0.5%) were non-oncology tests placed by 10 providers. The most common conforming tests were in prostate (55.9%), breast (14.1%) and pancreatic cancers (6.6%), while the most common non-conforming tests with a documented indication were ordered in patients with family history of malignancy (12.1%), lung cancer (9.7%) and melanoma (4.8%). Conclusions: Our findings demonstrate that high fidelity to germline cancer genetic testing guidance by oncology providers is achievable under the mainstream model with 94.7% of testing conforming to guidance. This is critical to our understanding of the effectiveness of mainstreaming and the ability to replicate and scale-up mainstreaming in different health care settings.

Individuals with germline BRCA1 or BRCA2 pathogenic variants (PV) may struggle with risk management decision-making. Advancements in technology could provide more specific risk information to patients, but the impact of this information is unknown. The Clinical Application of Refined Risk Estimates Study is a two-arm randomized controlled trial in women with a BRCA1/BRCA2 PV. The primary objective was to determine whether genotype-informed personalized cancer risk estimates (GRE) compared with standard lifetime cancer risk estimates (SRE) decreased decisional conflict related to cancer risk management decision-making. Women were recruited following the disclosure of their PV results. Participants completed a baseline survey and were randomized 1:1 to receive a GRE or SRE. After receiving their results, participants completed a follow-up survey. Likert and continuous data measures were analyzed using linear regression. There were no differences in decisional conflict between study arms at follow-up. However, individuals in the SRE arm showed an increased need for personal structure compared with those in the GRE arm (P = 0.02). Compared with baseline, individuals within the SRE arm showed decreased decisional conflict (P = 0.003) and increased perceived stress (P = 0.02) at follow-up. A more personalized cancer risk estimate did not decrease decisional conflict in women with BRCA1/BRCA2 PVs. Future studies will determine whether a GRE affects actual decision-making behaviors. PREVENTION RELEVANCE:Women with a germline PV in BRCA1 or BRCA2 have significantly elevated risks of developing breast and ovarian cancers. This randomized controlled trial evaluates the impact of polygenic risk scores on decisional conflict related to breast and ovarian cancer prevention and risk management in those with BRCA1/BRCA2 PVs.

Figure S6. Second replicate of ChIP-sequencing and HOMER motif analysis showing retained binding of R181-mutants to ETS sites.

Table S8: Ingenuity Pathway Analysis of RNAseq data after 8hr of Nutlin-3a in HCT116 p53 R181H/- vs ± cells.

Abstract The p53 tumor suppressor binds DNA cooperatively as a tetramer, mediated by salt-bridge interactions between p53 residues E180 and R181 from 2 different p53 monomers. Variants at the R181 residue are one of the most identified TP53 pathogenic variants by germline genetic testing. We show that families with TP53 p.R181H and p.R181C variants have an attenuated cancer risk phenotype compared with patients with hotspot dominant-negative loss-of-function TP53 variants. Despite this phenotype, we find that p53 R181H and R181C variants have significantly reduced ability to bind to p53 promoter/enhancer target sequences and transactivate p53 target genes, similar to null variants. However, p53 R181H and R181C retain wild-type (WT) p53 structure and tetramerization. In addition, R181-mutant cells undergo apoptosis through WT p53 activity at the mitochondria. These results suggest that retention of transcription-independent p53 tumor suppressor function results in a reduced penetrance cancer risk syndrome in humans. Implications: We report the first separation of function DNA-binding domain p53 mutation that results in retention of transcription-independent p53 functions despite loss of p53 transactivation activity, resulting in a reduced penetrance phenotype.

Table S5: Kratz criteria assigned to carriers of R181, DN/LOF, and hypomorphic variants in the patient cohort.

Figure S1. TP53 c.542G>A;p.R181H and TP53 c.541C>T;p.R181C variants are the most common variants observed at Penn, DFCI, and HCI and confer the risk of various tumors.

Figure S5. RNA-sequencing and quantitative PCR of R181-mutant cell lines confirm reduced induction of p53 target genes.

Abstract TP53 is the most frequently mutated gene in cancer, and its encoded protein p53 has many tumor-suppressive functions. p53 primarily acts as a transcription factor and binds to target sites on DNA cooperatively as a tetramer. This cooperative binding is mediated by salt-bridge interactions between p53 residues E180 and R181 from two different p53 monomers. Variants at the R181 residue are one of the most identified TP53 pathogenic variants by germline genetic testing, however the mechanism by which these variants disrupt p53 tumor suppression is not understood. We show that families with TP53 p.R181H and p.R181C variants have an attenuated cancer risk phenotype compared to patients with hotspot loss of function TP53 variants. Despite this clinical phenotype, we find that p53 R181H and R181C variants have significantly diminished ability to transactivate a set of ∼300 known p53 target genes in CRISPR knock-in colorectal and breast cancer cell lines. This loss of transactivation ability does not occur through defects in p53 structure or oligomerization, but through reduced cooperative binding to p53 target sites on DNA as determined using fluorescence polarization assays on purified p53 proteins and using chromatin immunoprecipitation sequencing in R181-mutant cancer cells. Despite the complete loss of p53’s transcriptional function, R181 mutants retain some tumor suppressive function. Colony formation assays show efficient colony suppression by R181H and R181C, and injecting R181 knock-in cancer cells into the subcutaneous tissue of mice results in comparable tumor progression levels between R181H, R181C, and wild-type p53. Interestingly, we observe residual apoptotic activity in R181H and R181C mutant cells when treated with DNA-damaging agent 5-fluorouracil, despite the poor transactivation of p53’s proapoptotic targets. This suggests that the R181 mutants retain the p53 transcription-independent mechanism of apoptosis, where p53 goes to the mitochondria to induce apoptosis. Indeed, proximity ligation assays between p53 and mitochondrial BAK show that R181 mutants traffic to the mitochondria upon genotoxic stress. Our study elucidates p53 tumor suppressive activities that are lost versus retained by R181 variants, which is estimated to account for 0.5% of all p53 missense mutations. Citation Format: Renyta Moses, Alexandra Indeglia, Alison Schwartz-Levine, Ryan Hausler, Gregory Kelly, Sven Miller, Isabel Anez, Melissa Heller, Rosella Delgado, Caitlin Orr, Wendy Kohlmann, Anne Naumer, Jennie Vagher, Sophie H. Cahill, Luke D. Maese, John Karanicolas, Judy E. Garber, Maureen E. Murphy, Kara N. Maxwell. Variation at the R181 residue of p53 confers loss of p53 DNA binding cooperativity with the retention of mitochondrial-associated apoptosis [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 1 (Regular Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(7 Suppl):Abstract nr 603.