Genetic Insights Shape Personalized Breast Cancer Management, Yet Variant Classification Uncertainty Persists

Genetic risk models and variant classification tools are refining breast cancer risk assessment by identifying high-risk individuals for intensive surveillance or targeted therapies, while guiding personalized, less-aggressive management for moderate- and lower-risk carriers, though further research is needed.¹

These topics were explored by experts during an afternoon educational session at the San Antonio Breast Cancer Symposium, titled “Management for Predisposition Gene Carriers.”

How Are Genetic Risk Models Transforming Breast Cancer Risk Assessment?

Tuya Pal, MD, of Vanderbilt University Medical Center, provided an overview of genetic risk models and their role in predicting the presence of germline pathogenic or likely pathogenic variants. She highlighted various models, including PREMMplus, which estimates the likelihood of carrying variants in 19 genes, about half of which are breast cancer genes. Individuals with a PREMMplus score of 2.5% or greater should be considered for multi-gene panel testing, Pal noted.

In comparison, National Comprehensive Cancer Network guidelines recommend testing for high-penetrance breast cancer genes based on personal and family history, with a threshold of 5% or greater risk using various predictive models, from PREMMplus to Tyrer-Cuzick and beyond.

Looking specifically at breast cancer risk prediction, Pal emphasized that accurate risk estimation informs clinical management of patients. She defined low, moderate, and high levels as having a less than 20%, 20% to 40%, and greater than 40% lifetime risk of developing breast cancer, respectively. Moderate risk typically warrants high-risk screening, such as breast MRI, while high-risk patients should consider surgical risk reduction with mastectomy.

Pal noted that guidelines for interventions like MRI or risk-reducing mastectomy differ across countries and institutions and may be based on outdated or debated thresholds. She argued instead that 5- and 10-year risk estimates are often more accurate than lifetime risk.

Pal then compared major breast cancer risk models, including Gail, Breast Cancer Surveillance Consortium Risk Calculator, BOADICEA/CanRisk, and Tyrer-Cuzick. These models incorporate personal characteristics, hormonal and reproductive risk factors, personal breast cancer history, and family history. She underscored that their inclusion of genetic status and polygenic risk scores (PRS) can significantly alter risk estimates, sometimes over- or under-estimating individual risk.

Of these models, only BOADICEA/CanRisk and Tyrer-Cuzick incorporate genetic status, with BOADICEA appearing more accurate, especially when integrating polygenic risk scores. Pal also highlighted CanRisk, which runs on the BOADICEA backbone, describing it as a comprehensive tool that integrates genetic results, PRS, breast density, personal factors, and family history. However, she warned that all models are imperfect.

“[They] give us an idea of the risk compared to the general carrier population of BRCA1, or whatever, but we can’t use the risks exactly as they are,” Pal claimed.

She next discussed emerging complexities in risk prediction, highlighting the shift from a binary model, which checked for the presence or absence of a mutation, to a continuous risk framework. Pal explained that some BRCA1/2 variants previously labeled as pathogenic are now recognized as lower penetrance, creating interpretation challenges.

She stressed the importance of distinguishing hypomorphic variants, which are associated with reduced gene function, from reduced-penetrance pathogenic variants, which indicate reduced clinical risk, noting that clinical relevance does not necessarily predict treatment response.

Through case examples, Pal illustrated how risk models can overestimate risk when lower-penetrance variants are treated as typical mutations. She also showed how PRS, breast density, and family history can modify risk for moderate-penetrance genes such as ATM or CHEK2; in some families, these variants may behave as high-risk due to strong modifying factors.

Pal concluded that breast cancer risk prediction is becoming increasingly nuanced and individualized.

“As we refine risks, my hope is that we’re going to be able to make the bucket smaller and smaller so that we have more refined confidence intervals in estimating risks so we’ll become better at estimating risks, but I don’t think we’ll ever be perfect,” she said. “What’s becoming evident is that risk prediction and refinement is becoming increasingly complex, and communication strategies will be critical.”

What Challenges Remain in Understanding and Classifying Breast Cancer Gene Variants?

Fergus J. Couch, PhD, of Mayo Clinic, expanded on Pal’s overview with a primer on genetics and the challenges of classifying gene variants. He focused on high-risk genes (BRCA1, BRCA2, and PALB2), moderate-risk genes (CHEK2 and ATM), and ultra-rare genes just beginning to be characterized. Couch noted that mutations in these genes carry varying breast cancer risks, ranging from about 70% lifetime risk for BRCA1/2 alterations to around 20% for certain moderate-penetrance genes, with ultra-rare variants remaining poorly understood.

He outlined key types of genetic variants, including nonsense and frame shift mutations, which typically disrupt protein function and are highly penetrant. Additionally, consensus splice sites and nearby intronic variants can have unpredictable effects, while copy number variants have less well-characterized risk. However, missense variants, in particular, are often classified as variants of uncertain significance (VUS), leaving clinicians and patients without clear guidance.

“[VUS] are changes in the genes that were picked up in a germline test, and we have no clue what they’re telling us or what the impact on disease actually is,” Couch said. “It’s a major issue in the clinical community, and so a lot of what we do is really focused on trying to figure out which ones are real and which ones are not.”

He emphasized that VUS should not guide patient management, with clinical decisions instead relying on personal and family history unless a variant is reclassified based on robust evidence. Typical methods to interpret VUS include case/family segregation data, population frequency, computational predictions, and functional assays. Following American College of Medical Genetics (ACMG) guidelines, these methods are all incorporated to assign pathogenicity scores, which predict the likelihood that a genetic variant causes disease.

Couch highlighted a major advancement in VUS identification: large-scale high-throughput functional studies using CRISPR/Cas9 genome editing in haploid cell systems. These allow every possible variant in a gene region to be tested for effects on cell viability and protein function.

Recent studies graded thousands of variants based on functional scores, effectively separating pathogenic variants from benign synonymous changes. Notably, 14% of missense variants in a key BRCA2 domain behaved like strong pathogenic alleles despite most appearing benign. Therefore, Couch noted that these results were much more nuanced than previous knowledge from clinical observation alone.

Incorporating functional assay results into the ACMG classification workflow dramatically reduces the number of unresolved VUS, he claimed. In Couch’s example, over 90% of tested variants could be confidently classified, leaving only a small fraction remaining ambiguous. However, he cautioned that results can differ between experiments or labs, with discrepancies arising in about 20% to 25% of variants, making cross-validation essential.

“There’s really rapid movement in this field, but I think, as I said, we need validation of a lot of these things, and then we need to run into these variant curation expert panels to classify all those variants and make sure that they’re done correctly,” Couch concluded.

How Is Hereditary Breast Cancer Managed?

Building on the earlier presentations, Stephanie M. Wong, MD, MPH, of McGill University, reviewed the evolution and advances in hereditary breast cancer management, first highlighting major advances in caring for BRCA/1/2 and PALB2 carriers. She explained that BRCA1/2 carriers with newly diagnosed breast cancer may pursue either breast-conserving therapy or bilateral mastectomy.

Although retrospective studies have historically shown no overall survival advantage for mastectomy over breast conservation, Wong noted that recent findings suggest a potential survival benefit bilateral mastectomy, but only in carriers under the age of 40. She also emphasized that contralateral breast cancer risk is strongly age- and gene-dependent, and because overall survival differences remain inconsistent, guidelines continue to encourage individualized counseling and patient preference in surgical decision-making.

Regarding systemic therapy, Wong highlighted that BRCA-associated tumors exhibit homologous recombination repair deficiency, making them highly responsive to DNA-damaging agents and poly (ADP-ribose) polymerase (PARP) inhibitors. Olaparib (Lynparza; AstraZeneca and Merck) and talazoparib (Talzenna; Pfizer) have demonstrated significant benefit in metastatic disease and, through the OlympiA trial (NCT02032823), in high-risk early-stage BRCA-associated breast cancer.² Evidence also supports PARP inhibitor use for PALB2-mutated metastatic breast cancer.¹ Based on emerging data, Wong added that combinations of PARP inhibitors with immunotherapy appear safe.

Wong next addressed management considerations for moderate-penetrance carriers, such as ATM and CHEK2. She noted that ATM and CHEK2 carriers generally have lower contralateral breast cancer risk, and most current guidelines do not recommend routine risk-reducing mastectomy for these patients. Instead, decisions should be driven by family history, personal risk factors, and patient preferences. Individualized counseling based on family history is suggested. Additionally, adjuvant radiation is considered safe for ATM carriers based on growing evidence showing low toxicity.

Looking ahead, Wong described ongoing research aimed at refining treatment management, including endocrine prevention studies in moderate-penetrance carriers and investigations of delayed oophorectomy strategies in BRCA carriers. Overall, she acknowledged the “tremendous advancements in the field of hereditary breast cancer, in large part because of contributions of researchers, many of whom are in this room or on this panel, and a very engaged patient population.”

References

1. Domchek S, Pal T, Couch FJ, Wong SM, Wedley-Majsiak B. Management for predisposition gene carriers. Presented at: San Antonio Breast Cancer Symposium 2025; December 9-12, 2025; San Antonio, Texas.
2. Olaparib as adjuvant treatment in patients with germline BRCA mutated high risk HER2 negative primary breast cancer (OlympiA). ClinicalTrials.gov. Updated August 23, 2024. Accessed December 11, 2025. https://clinicaltrials.gov/study/NCT02032823