Association between polygenic risk and survival in breast cancer patients

Danielle E. Kurant, Stefan Groha, Yi Ding, Chris German, Wei Wang, Julie M. Granka, Michael V. Holmes, Suyash Shringarpure, Alexander GusevmedRxiv 2024 · DOI

Abstract

Polygenic risk scores (PRS) estimate an individual’s germline genetic predisposition to disease by aggregating the effects of thousands of genetic variants. While PRS are primarily used for cancer risk screening, their influence on survival post-diagnosis is less understood. This study analyzed 23,770 European cancer patients from the Dana-Farber Cancer Institute (DFCI) Profile cohort to establish whether PRS for cancer and other traits influence patient outcomes. We identified a significant association between high breast cancer PRS and improved survival (HR = 0.89, p = 1.50×10⁻⁴), suggesting that individuals at high genetic risk develop less aggressive tumors. This was supported by a lower prevalence of somatic TP53 mutations in high-PRS individuals. The association was strongest in ER-negative tumors (HR = 0.78) and was not observed in other cancer types. Conversely, PRS for type 2 diabetes and bipolar disorder were associated with shorter survival. These findings suggest that germline risk provides novel insights into tumor development and prognosis.

Introduction

Polygenic Risk Scores (PRS) have become a cornerstone of genomic research, offering a way to quantify an individual's inherited susceptibility to various conditions. In oncology, the focus has largely been on using these scores to identify who is most likely to develop cancer. However, the impact of these inherited factors on the clinical course of the disease after diagnosis remains a critical gap in our knowledge.

Does a high genetic predisposition lead to a more aggressive form of cancer, or does it fundamentally change how the tumor evolves? This study explores the relationship between germline risk and patient survival, providing a potential biological explanation for why some patients fare better than others.

Methods and Data

The researchers conducted a large-scale survival analysis using the Dana-Farber Cancer Institute (DFCI) Profile cohort, which included 23,770 patients of European ancestry.

  • PRS Calculation: 209 different PRS were evaluated, covering cancer risks, autoimmune diseases, and metabolic traits. Most scores were trained using data from the 23andMe Research Cohort.
  • Survival Modeling: The team used Cox Proportional-Hazards models to analyze the association between these scores and overall survival across multiple clinical index dates (diagnosis, sequencing, and start of palliative care).
  • Somatic Integration: A key feature of the study was the integration of germline data with somatic (tumor) sequencing to identify mutations in genes like TP53.
  • Validation: Findings were replicated using independent weights from the PGS Catalog to ensure the results were not specific to the initial models.

The "Polygenic Two-Hit" Model

The most striking result of the study is a counter-intuitive phenomenon: patients with a higher inherited risk for breast cancer actually tend to survive longer. Specifically, each standard deviation increase in breast cancer PRS was associated with an 11% reduction in the risk of death (HR = 0.89).

The authors propose a "polygenic two-hit model" to explain this. If an individual's germline genetics provide a significant "head start" toward developing cancer, the resulting tumor may not need to acquire as many high-impact somatic mutations to grow. Consequently, these tumors are biologically less aggressive and more treatable than tumors that develop in individuals with low genetic risk, which must undergo more extreme evolutionary changes to manifest.

Key Insights

TP53 as a Marker of Aggression

Supporting the two-hit model, high-PRS individuals were 15% less likely to harbor somatic TP53 mutations (OR = 0.85), which are typically associated with aggressive, hard-to-treat cancers. When TP53 mutations did occur in these patients, they were often subclonal, suggesting they appeared later in the tumor's development rather than acting as primary drivers.

The ER-Negative Effect

While Estrogen Receptor (ER)-negative breast cancer generally carries a worse prognosis, the protective effect of a high PRS was actually more pronounced in this subgroup. The Hazard Ratio dropped to 0.78, indicating a 22% reduction in the risk of death for high-PRS patients with ER-negative tumors.

Specificity to Breast Cancer

Interestingly, this "high risk = better survival" link was unique to breast cancer. It was not observed in lung, prostate, or ovarian cancers. This suggests that the interplay between inherited risk and tumor evolution is highly dependent on the specific biological context of the organ involved.

Impact of Non-Cancer PRS

While breast cancer risk was protective, the study found that PRS for Type 2 diabetes, bipolar disorder, and pancreatitis were associated with shorter survival (HR ≈ 1.04). This likely reflects the negative impact of systemic comorbidities on a patient's overall resilience and response to cancer therapy.

Limitations and Future Work

Despite the large sample size, the study was restricted to individuals of European ancestry, meaning the results may not yet be generalizable to other populations. Additionally, as the DFCI is a major referral center, the cohort may represent more complex cases than the general population.

Future research will need to determine if these findings can be used to refine clinical staging or if high-PRS patients might benefit from different therapeutic intensities. The researchers also suggest that further investigation into "digital pathology" could help clarify if high PRS is identifying a specific spectrum of tumor biology that current testing misses.

Conclusion

This study demonstrates that the genetic variants we inherit do more than just predict our risk of getting sick; they may also influence how we recover. By showing that high polygenic risk is associated with less aggressive breast cancer and improved survival, this research provides a new lens through which to view cancer prognosis and precision medicine.

Summary updated

Abstract

Abstract Polygenic risk scores (PRS) estimate an individual’s germline genetic predisposition to a quantitative trait and/or risk of disease. Several PRS have been developed for cancer risk with the goal of improved risk screening. However, little is known about the association with subsequent outcomes for individuals who develop cancer. Here, we sought to establish whether PRS for cancer risk and other common traits may influence survival for patients with cancer. We conducted a PRS survival analysis using 23,770 European cancer patients from the Dana-Farber Cancer Institute Profile cohort. We identified an association between PRS for breast cancer risk and longer patient survival (HR = 0.89, p = 1.50×10 -4 , <5% FDR), implying that individuals at high genetic risk had better outcomes. High PRS individuals were also significantly less likely to harbor somatic TP53 mutations, consistent with having less aggressive tumors. This association persisted when including tumor grade and became more protective when restricting to ER-negative tumors (HR = 0.78, p = 1.69×10 -4 ). Potential confounders such as hormone receptor status, age, grade, stage, and ER-targeted therapy did not fully explain this association, nor was there statistical evidence of index event bias at individual variants. We did not observe significant associations between cancer risk and survival for other cancers, suggesting that this mechanism may be largely unique to breast cancer. However, we did observe associations between shorter survival and type 2 diabetes, bipolar, and pancreatitis PRS (1% FDR). These findings suggest that higher germline risk may predispose individuals to less aggressive breast cancer tumors and provide novel insights into breast cancer development and prognosis.
BibTeX
@misc{kurant2024association,
  author = {Danielle E. Kurant and Stefan Groha and Yi Ding and Chris German and Wei Wang and Julie M. Granka and Michael V. Holmes and Suyash Shringarpure and Alexander Gusev},
  title = {{Association between polygenic risk and survival in breast cancer patients}},
  year = {2024},
  doi = {10.1101/2024.10.31.24315531},
  howpublished = {medRxiv},
}
Figure