Study validates a breast cancer risk score in the Brazilian population and reinforces pathways for precision medicine in developing countries 

Published in Diagnostics, the study involved researchers from the D’Or Institute for Research and Education (IDOR) and investigated whether polygenic risk scores developed using European populations can accurately predict breast cancer risk in Brazilian women, a population characterized by extensive genetic admixture. 

Polygenic risk: when many small genetic factors matter more than a single one 

Breast cancer is the most common cancer among women worldwide and in Brazil, and it is estimated that one in eight women will develop breast cancer during their lifetime. Part of the risk is related to lifestyle and environmental factors, while another part is genetic. 

Although some rare mutations in genes such as BRCA1 and BRCA2 greatly increase the likelihood of developing the disease, they account for a relatively small proportion of cases. Most of the genetic predisposition to breast cancer comes from what is known as polygenic risk. This means that risk is not concentrated in a single gene, but rather in the combined effect of hundreds or thousands of common genetic variants. Individually, each variant has only a small impact, but together they can significantly alter the probability of developing the disease. 

A polygenic risk score, or PRS, is a way of converting this accumulation of small genetic variations into a numerical value. This number indicates whether a person falls below, near, or above the average population risk. The challenge is that most of these models were developed primarily using European populations. In a highly admixed country such as Brazil, with European, African, Indigenous, and Asian ancestry, an important question arises: do these genetic scores work in the same way? 

More than 15,000 Brazilians analyzed 

To answer this question, researchers evaluated genetic data from 15,490 Brazilians. The sample included 6,362 women with a history of breast cancer, while the remaining participants were adults without a diagnosis of the disease and served as a control group. 

Genetic material was analyzed using whole-exome sequencing, a technique that examines the regions of DNA responsible for coding proteins. Because not all variants of interest are directly captured through this type of sequencing, researchers used a statistical strategy known as imputation. In practice, this approach relies on large international databases, such as the 1000 Genomes Project, to infer variants that were not directly observed in the sequencing data. 

Using these data, four different polygenic risk score models were tested: Broad, PRS313, PRS3820, and UKBB. 

Researchers also estimated the genetic ancestry of each participant to evaluate whether the performance of the scores differed among individuals with predominantly European, non-European, or East Asian ancestry. 

Can Brazilian admixture reduce the effectiveness of genetic tests developed from European populations? 

Before analyzing the impact of polygenic scores, researchers excluded women carrying rare, high-impact mutations, such as those found in BRCA1 and BRCA2, which independently confer a substantially increased risk of breast cancer. The goal was to specifically assess the cumulative effect of common variants. 

The results confirmed that genetic admixture does affect the performance of the models tested, producing weaker effects in the Brazilian population than those observed in the original European studies. This occurs because genetic frequencies and inheritance patterns differ in admixed populations. Based on their findings, the authors concluded that the Broad and UKBB models are not suitable for assessing population risk in Brazil. 

The good news is that the other two models, particularly PRS3820, still performed very well in identifying breast cancer risk among Brazilian women, regardless of whether their ancestry was predominantly European, African, or Asian. 

Validation of this score in Brazil is especially important because it demonstrates that the test’s ability to identify risk in European women can be translated with very similar effectiveness to Brazilian women from different ancestral backgrounds. This means, for example, that being among the 10% highest-risk individuals according to the PRS3820 score, whether Brazilian or European, corresponds to a meaningful increase in breast cancer risk that should not be overlooked in clinical practice. 

Another positive finding was that exome sequencing data combined with imputation may represent a more affordable alternative to whole-genome sequencing for calculating polygenic risk scores, potentially facilitating their broader implementation. 

A step toward integrating rare genetics and cumulative risk 

The results reinforce that both rare high-impact mutations and polygenic risk contribute to hereditary breast cancer risk in Brazil. Approaches that integrate these two dimensions may allow for more refined risk stratification. 

However, before these scores can be incorporated into routine clinical practice, prospective studies will still be needed to determine how best to use them in screening programs, such as defining earlier screening initiation or personalized mammography intervals. 

Even so, by demonstrating that models such as PRS3820 maintain meaningful predictive performance in a highly admixed population, the study represents a concrete advance for precision medicine in Brazil, enabling higher-risk patients to be identified and treated earlier, thereby improving the likelihood of positive outcomes. 

Written by Maria Eduarda Ledo de Abreu. 

About IDOR 

Founded in 2010, the D’Or Institute for Research and Education (IDOR) is a Brazilian non-profit institution dedicated to research, education, and innovation in healthcare. Operating nationally and internationally, the Institute conducts high-quality studies across multiple areas of biomedical science, trains professionals, and contributes to scientific advances that improve public health. Its main sponsor is Rede D’Or, Brazil’s largest integrated healthcare network. More information: www.idor.org.