Triglyceride Polygenic Score Identifies Differential Bleeding and Cardiovascular Risk with Aspirin in Primary Prevention

This study demonstrates that a triglyceride-related polygenic score can stratify older adults in primary prevention, identifying those with high genetic risk who are harmed by aspirin (increased bleeding without cardiovascular benefit) versus those with low genetic risk who are protected from both bleeding and cardiovascular events.

The Gist

Imagine you are a doctor trying to decide whether to give an older patient a daily aspirin to prevent heart attacks. For a long time, the rule was "better safe than sorry." But recently, doctors realized that for many older adults, the risk of the aspirin causing a dangerous bleed (like a brain hemorrhage or stomach bleed) is actually higher than the benefit of preventing a heart attack. It's like using a sledgehammer to crack a nut; sometimes the tool causes more damage than the problem it's trying to fix.

So, the big question became: Is there a way to tell exactly who should take aspirin and who should avoid it?

This study says: Yes, and the answer might be written in your DNA.

Here is the story of the study, explained simply:

The "Genetic ID Card"

Think of your genes as a massive instruction manual for your body. In this study, the researchers looked at a specific set of instructions related to triglycerides. You can think of triglycerides as a type of fat floating in your blood. Usually, we worry about having too much of this fat because it clogs arteries.

The researchers created a "Triglyceride Polygenic Score" (PGS). Imagine this score as a genetic ID card that predicts how your body naturally handles these fats, regardless of what you eat or how much you exercise.

The Two Groups: The "Fragile" vs. The "Resilient"

The researchers took 13,571 older adults from a big trial (called ASPREE) and looked at their genetic ID cards. They split them into groups based on their scores. They found two very different reactions to aspirin:

1. The "Fragile" Group (Low Triglyceride Score)

• Who they are: These people have a genetic makeup that naturally keeps their triglyceride levels low.
• What happened: When they took aspirin, it was like pouring gasoline on a fire. Their risk of serious bleeding skyrocketed (more than double the risk compared to those who took a placebo).
• The Result: For this group, aspirin was dangerous. It didn't stop heart attacks, but it significantly increased the chance of a major bleed.
• The Metaphor: Imagine a house with very thin, fragile walls. If you try to fix a small crack with a heavy hammer (aspirin), you might accidentally knock the whole wall down (bleeding).

2. The "Resilient" Group (High Triglyceride Score)

• Who they are: These people have a genetic makeup that naturally keeps their triglyceride levels higher.
• What happened: When they took aspirin, it was like putting a shield on a fortress. They had lower risks of bleeding and lower risks of heart attacks or strokes.
• The Result: For this group, aspirin was a hero. It protected their hearts without causing harm.
• The Metaphor: Imagine a house with thick, reinforced concrete walls. You can use the heavy hammer (aspirin) to fix the cracks, and the walls stay standing strong.

Why This Matters

For years, doctors have treated all older adults the same way: "Don't take aspirin unless you already have heart disease." This study suggests that one size does not fit all.

• The Old Way: A blunt instrument. We either give it to everyone or no one, guessing who will get hurt.
• The New Way (Precision Medicine): We check the "Genetic ID Card" first.
  • If your card says "Fragile," we say, "No aspirin for you, it's too risky."
  • If your card says "Resilient," we say, "Go ahead, this could save your life."

The Bottom Line

This research is like finding a genetic filter. It helps us separate the people who will be harmed by aspirin from the people who will be saved by it.

While this study is a "preprint" (meaning it's a new discovery that hasn't been fully double-checked by other scientists yet), it offers a hopeful future. Instead of guessing, we might soon be able to look at a patient's DNA, see their "Triglyceride Score," and prescribe aspirin only to those who will truly benefit, sparing others from unnecessary danger.

In short: Your genes might be the key to knowing if aspirin is your best friend or your worst enemy.

Technical Summary

1. Problem Statement

Low-dose aspirin is no longer routinely recommended for the primary prevention of cardiovascular disease in older adults because the risk of major bleeding often outweighs the cardiovascular benefits in the general population. Large randomized trials (ARRIVE, ASCEND, ASPREE) have demonstrated a net unfavorable benefit-to-harm ratio. However, these trials treated participants as a homogeneous group, failing to account for interindividual variability in aspirin response driven by genetic factors. There is a critical need to identify specific subgroups where aspirin might provide a net benefit or, conversely, where the bleeding risk is prohibitively high, to enable precision medicine in primary prevention.

2. Methodology

The study utilized a post-hoc genetic analysis of the ASPREE (Aspirin in Reducing Events in the Elderly) trial, a large randomized, placebo-controlled trial conducted in Australia and the US.

• Study Population: 13,571 participants (median follow-up 4.6 years) aged ≥70 (≥65 for US minorities) without prior cardiovascular disease, dementia, or physical disability.
• Intervention: Randomization to 100 mg daily enteric-coated aspirin or placebo.
• Genetic Screening:
  • Researchers screened 572 Polygenic Scores (PGSs) derived from the PGS Catalog, covering cardiovascular, hematologic, inflammatory, and lipid traits.
  • PGSs were generated using PLINK 1.9 based on GWAS data (European ancestry focus) and filtered for quality control (SNP availability, distributional normality).
• Statistical Analysis:
  • Primary Model: Cox proportional-hazards models tested for interaction effects between aspirin allocation and each PGS on the outcome of major bleeding.
  • Covariates: Models were adjusted for age, sex, the first 10 genetic principal components, and subsequently for clinical factors (smoking, BMI, hypertension, diabetes, CKD, statin use).
  • Correction: P-values for interactions were Bonferroni-corrected to account for multiple testing.
  • Stratification: Significant PGSs were analyzed by quintiles to compare aspirin vs. placebo effects on Major Bleeding and Major Adverse Cardiovascular Events (MACE).
  • Exploratory Analysis: Baseline measured fasting triglyceride levels were correlated with the significant PGS to validate findings.

3. Key Contributions

• Systematic PGS Screening: This is one of the first studies to systematically screen hundreds of polygenic scores to identify genetic modifiers of aspirin-induced bleeding risk.
• Identification of a Specific Genetic Modifier: The study identified a triglyceride-related Polygenic Score (PGS003144) that significantly interacts with aspirin use.
• Divergent Risk Stratification: The study demonstrates that the same drug (aspirin) produces diametrically opposite risk profiles depending on an individual's genetic predisposition to triglyceride levels.
• Validation with Phenotypic Data: The genetic findings were corroborated by exploratory analysis of measured baseline triglyceride levels, showing a consistent pattern of risk modification.

4. Key Results

The analysis revealed a significant interaction between the Triglyceride PGS (PGS003144) and aspirin allocation for major bleeding ($P_{interaction} = 5.9 \times 10^{-5}$; Bonferroni-adjusted $P = 0.034$).

A. Lowest Triglyceride PGS Quintile (Q1):

• Bleeding Risk: Aspirin use was associated with a markedly increased risk of major bleeding compared to placebo (Hazard Ratio [HR] 2.28; 95% CI 1.45–3.58).
  • Gastrointestinal bleeding HR: 3.17.
  • Intracranial bleeding HR: 4.10.
• Cardiovascular Benefit: Aspirin provided no reduction in MACE (HR 1.04; 95% CI 0.67–1.62).
• Conclusion for Q1: Net harm; aspirin increases bleeding without cardiovascular benefit.

B. Highest Triglyceride PGS Quintile (Q5):

• Bleeding Risk: Aspirin use was associated with a lower risk of major bleeding compared to placebo (HR 0.62; 95% CI 0.38–0.97).
• Cardiovascular Benefit: Aspirin was associated with a significant reduction in MACE (HR 0.66; 95% CI 0.44–0.99).
• Conclusion for Q5: Net benefit; aspirin reduces both bleeding and cardiovascular events.

C. Measured Triglycerides:

• A moderate correlation existed between measured triglycerides and the PGS ($r=0.30$).
• Participants in the lowest measured triglyceride quintile assigned to aspirin also showed increased bleeding risk (HR 1.68), supporting the genetic findings, though the effect size was smaller than the genetic stratification.

5. Significance and Implications

• Precision Medicine in Primary Prevention: The findings challenge the "one-size-fits-all" approach to aspirin prescribing. They suggest that genetic profiling could identify older adults who are likely to suffer net harm (low triglyceride PGS) versus those who may derive net benefit (high triglyceride PGS).
• Biological Mechanism: The results suggest a complex relationship between lipid metabolism, vascular integrity, and aspirin pharmacodynamics.
  • Low Triglyceride PGS: May reflect underlying vascular fragility or hemostatic imbalances that aspirin exacerbates.
  • High Triglyceride PGS: May reflect a state of enhanced platelet reactivity where aspirin provides a protective effect against thrombosis without increasing bleeding risk.
• Clinical Guidance: While the study is hypothesis-generating and requires validation in diverse populations, it provides a strong rationale for future trials to stratify participants by lipid-related genetic scores. This could optimize the benefit-to-harm ratio of aspirin, potentially reversing current guidelines that broadly discourage its use in the elderly.
• Limitations: The study is post-hoc, the cohort is predominantly of European ancestry (limiting generalizability), and the specific causal biological mechanisms remain to be elucidated.

Conclusion: A triglyceride-related polygenic score effectively stratifies older adults into subgroups with divergent bleeding and cardiovascular responses to aspirin, offering a promising pathway for individualized primary prevention strategies.