Genomic ascertainment of PALB2-related cancer predisposition

This study demonstrates that genomic ascertainment of heterozygous *PALB2* pathogenic variants in large population cohorts reveals significantly elevated risks for breast and pancreatic cancers, as well as increased all-cause mortality, with risks further modified by family history but generally lower than estimates derived from familial ascertainment.

The Gist

Imagine your body is a massive, bustling city. Every day, billions of construction crews (your cells) are building, repairing, and maintaining the city's infrastructure. To keep everything safe, the city has a specialized "Emergency Repair Team" called PALB2. Its job is to fix broken DNA (the city's blueprints) whenever a storm or accident damages them.

Sometimes, a person is born with a broken blueprint for this repair team. They have a PALB2 mutation. It's like having a repair crew that is missing a few key tools. They can still work, but they aren't as efficient as a fully equipped team. This makes the city (the body) more vulnerable to "construction errors" that can turn into dangerous structures, or cancers.

The Big Question: How Dangerous is the Broken Blueprint?

For a long time, doctors only knew how dangerous this broken blueprint was by looking at families where cancer was already a big problem. It's like trying to figure out how bad a storm is by only looking at the houses that were already destroyed. This method tends to make the storm look scarier than it might be for the average person.

This new study wanted to know: What if we find people with this broken blueprint before they get sick? What is their actual risk?

To answer this, the researchers acted like "genetic detectives." They didn't wait for people to get sick. Instead, they looked at the DNA of 636,000 volunteers from two huge databases (the UK Biobank and Geisinger MyCode). They found about 1,000 people who carried this specific broken blueprint, even though many of them didn't know it yet.

What They Discovered: The "Risk Map"

Here is what the study found, translated into everyday terms:

1. The "All-Cancer" Warning Light
People with this mutation have a higher chance of getting any type of cancer compared to people without it.

• The Analogy: Imagine your city has a slightly higher chance of catching fire, not just in one specific building, but in various districts.
• The Numbers: The risk of getting cancer was about 1.7 to 3.6 times higher than for someone without the mutation.

2. The "Breast and Pancreas" Hotspots
While the risk goes up for many cancers, two areas are the most critical:

• Breast Cancer: For women, the risk is significantly higher. By age 80, about 27% to 38% of women with this mutation will develop breast cancer.
  • Context: If you look at families with a lot of cancer history, the risk looks like it could be as high as 80%. But in the general population (where we found these people randomly), the risk is lower, though still very serious.
• Pancreatic Cancer: The risk for this is also elevated, though it's rarer.

3. The "Family History" Multiplier
The study found that if you have the mutation and your family has a history of cancer, your risk goes up even more.

• The Analogy: Think of the mutation as a weak foundation for a house. If you also live in a flood zone (family history), the house is in much more danger than if you just had the weak foundation but lived on a hill.

4. The "Mortality" Reality Check
This was a sobering finding. People with this mutation who did get cancer had a harder time surviving it than people without the mutation who got the same cancer.

• The Analogy: It's like two cars getting into a crash. The car with the broken repair team (the mutation) might have more damage or be harder to fix, leading to a worse outcome.

Why This Matters for You

This study is a game-changer because it changes how we view "genetic screening."

• Old Way: "We only test you if your mom or sister had cancer." (This finds the worst cases).
• New Way: "We can test everyone's DNA, find the broken blueprint early, and give them a personalized safety plan."

The Takeaway:
Finding a PALB2 mutation doesn't mean you will get cancer, but it does mean your "Emergency Repair Team" needs extra help.

• Surveillance is Key: If you have this mutation, you need to be more vigilant. Think of it as installing extra smoke detectors and fire extinguishers in your house.
• Early Detection: Because the risk is higher, doctors might suggest starting cancer screenings (like mammograms) earlier or doing them more often.
• Lifestyle: Knowing your risk can motivate healthier choices to keep your city safe.

In short, this paper tells us that while the "broken blueprint" is serious, knowing about it before disaster strikes gives us the power to protect the city better than ever before. It's not a death sentence; it's a warning label that allows us to take action.

Technical Summary

1. Problem Statement

While the cancer risks associated with germline PALB2 pathogenic variants (P/LP) are well-established in families with a history of cancer (phenotypic ascertainment), there is a significant knowledge gap regarding the risk magnitude when these variants are identified through genomic ascertainment (e.g., opportunistic screening, secondary findings, or incidental discovery).

• The Gap: Traditional risk estimates often suffer from ascertainment bias, potentially overestimating penetrance. It is unknown how PALB2 risks translate to general population cohorts where individuals are identified by genotype rather than phenotype.
• Specific Unknowns: The study aims to quantify cancer incidence, age-specific penetrance, and all-cause mortality in PALB2-heterozygotes identified via genomic screening, and to determine how family history modifies these risks.

2. Methodology

The researchers conducted a large-scale, retrospective case-control study using two distinct population-based biobanks:

• Cohorts:
  • UK Biobank (UKBB): $n = 469,580$ participants.
  • Geisinger MyCode: $n = 167,050$ participants (adults $\ge$ 18 years).
• Variant Identification:
  • Utilized exome sequencing data.
  • Filtered for high-quality germline variants (Genotype Quality $\ge$ 30, Depth $\ge$ 5, Allelic Balance 0.3–0.7).
  • Classified variants as Pathogenic (P) or Likely Pathogenic (LP) using AutoGVP and InterVar, with manual review by a medical geneticist.
  • Cases: Individuals with heterozygous PALB2 P/LP variants.
  • Controls: Non-carriers.
• Data Sources:
  • Cancer diagnoses linked from national cancer registries and electronic health records (EHR).
  • Vital status and mortality data from death registries.
• Statistical Analysis:
  • Odds Ratios (OR): Multivariable logistic regression adjusted for birth year, sex, smoking, and BMI to assess cancer prevalence.
  • Hazard Ratios (HR): Cox proportional hazards models (using age as the time scale) to assess time-to-event (cancer diagnosis and all-cause mortality).
  • Penetrance: Age-specific cumulative incidence estimated via Kaplan-Meier methods.
  • Stratification: Analyses were stratified by family history of cancer (where data was available, primarily in UKBB).

3. Key Contributions

• Genomic vs. Phenotypic Risk Quantification: Provides the first large-scale estimates of PALB2 cancer risk derived specifically from genomic ascertainment, allowing for a direct comparison with familial risk estimates.
• Founder Variant Analysis: Identified a significant founder effect in the UKBB cohort driven by the p.Trp1038Ter variant, allowing for variant-specific risk stratification.
• Mortality Impact: Quantified the impact of PALB2 status on all-cause mortality and post-cancer survival, a metric previously unexplored in genomically ascertained cohorts.
• Broad Spectrum Analysis: Evaluated risks across multiple organ systems, including rare associations like neuroendocrine tumors and skin cancers, beyond the established breast and pancreatic risks.

4. Key Results

Prevalence and Demographics

• Prevalence: PALB2 P/LP variant frequency was 1:571 in UKBB and 1:940 in MyCode.
• Founder Effect: The higher prevalence in UKBB was driven by an ~80-fold excess of the p.Trp1038Ter variant (a common British Isles founder variant), which was absent in the MyCode cohort.

Cancer Risk (Odds Ratios & Hazard Ratios)

• Overall Cancer: PALB2-heterozygotes had significantly increased odds of any cancer in both cohorts.
  • Adjusted HR for any cancer: 1.74 (MyCode) and 1.74 (UKBB).
• Breast Cancer:
  • Significantly elevated risk in females.
  • Adjusted HR: 1.85 (MyCode) and 3.56 (UKBB).
  • Penetrance by Age 80: ~38% (MyCode) and ~27% (UKBB) for females. This is notably lower than the ~80% penetrance often cited in familial studies.
  • DCIS: Significant excess risk of Ductal Carcinoma In Situ (DCIS) observed in both cohorts.
• Pancreatic Cancer:
  • Significant excess risk observed, primarily driving the "digestive organ" signal.
  • Risk divergence from controls began around age 70 in UKBB.
• Other Cancers:
  • Skin: Excess risk of melanoma and non-melanoma skin cancers observed in UKBB.
  • Neuroendocrine: Elevated odds of neuroendocrine tumors (though numbers were modest).
  • No Association: No consistent evidence of increased risk for colorectal or prostate cancers.

Family History Modification

• In UKBB, female PALB2-heterozygotes with a family history of breast cancer showed a non-significant increase in risk compared to those without a family history, though the trend suggested higher risk.

Mortality

• All-Cause Mortality: Significantly increased in PALB2-heterozygotes compared to controls.
  • Adjusted HR: 1.67 (MyCode) and 1.61 (UKBB).
  • Divergence in mortality curves occurred around age 55 (MyCode) and 65 (UKBB).
• Post-Diagnosis Survival: Among individuals who developed cancer, PALB2-heterozygotes had significantly reduced overall survival compared to non-carriers with cancer. This was observed for all cancers and specifically for female breast cancer in UKBB.

5. Significance and Clinical Implications

• Risk Management: The study confirms that PALB2 carriers identified via genomic screening face elevated risks for breast, pancreatic, and other cancers, necessitating clinical surveillance. However, the absolute risk (penetrance) is lower than estimates derived from high-risk familial cohorts.
• Surveillance Bias: The lower penetrance in genomic cohorts compared to familial cohorts may be attributed to the "healthy volunteer effect" in biobanks and the lack of ascertainment bias (i.e., these individuals were not selected because they already had cancer).
• Therapeutic Implications: The finding of reduced survival in PALB2-carriers with cancer suggests that genotype status may influence disease aggressiveness or treatment response. The authors note that many carriers in these cohorts may have been treated without knowledge of their genotype, potentially missing opportunities for targeted therapies (e.g., PARP inhibitors).
• Future Directions: The study highlights the need for variant-specific risk estimates (e.g., distinguishing p.Trp1038Ter from other variants) and further investigation into associations with skin and neuroendocrine cancers.

Conclusion: Genomic ascertainment of PALB2 variants reveals a clear, elevated risk profile for multiple cancers and increased mortality, but with lower absolute penetrance than previously thought based on family history alone. These findings are critical for counseling individuals identified through population screening and for refining risk management guidelines.