#290 ‒ Liquid biopsies for early cancer detection, the role of epigenetics in aging, and the future of aging research | Alex Aravanis, M.D., Ph.D.

Feb 19, 2024 Episode Page ↗
Overview

Peter Attia and Alex Aravanis, CEO of Moonwalk Biosciences, discuss liquid biopsies for early cancer detection, tracing genome sequencing's evolution and Grail's Galleri test. They also delve into epigenetics, DNA methylation, and its potential role in cellular reprogramming and reversing aging.

At a Glance
12 Insights
2h 5m Duration
13 Topics
9 Concepts

Deep Dive Analysis

Alex Aravanis's Engineering Background and Early Research

Human Genetics Primer and Evolution of DNA Sequencing

The Advent and Evolution of Liquid Biopsies

Cell-Free DNA and Incidental Findings in Prenatal Testing

Developing a Universal Blood Test for Cancer Detection

GRAIL's Multi-Cancer Screening Test Development (CCGA Study)

Understanding DNA Methylation and its Role

Optimizing Cancer Detection with Methylation Analysis

Importance of Sensitivity, Specificity, and Predictive Values

Performance of the GRAIL Galleri Test

Does Early Cancer Detection Improve Overall Survival?

The Role of Epigenetics in Aging Biology

Cellular Reprogramming and the Future of Rejuvenation

DNA Sequencing

The process of reading out the specific ATCG code of an individual's DNA. Early methods involved cutting large sections and sequencing them, while later 'shotgun sequencing' involved breaking the entire genome into small fragments and computationally reassembling them, leading to massive cost reductions and speed improvements.

Cell-Free DNA (cfDNA)

Fragments of DNA found in the bloodstream outside of cells, typically around 160 base pairs long. This DNA is protected by nucleosomes and is released into circulation primarily from dying cells due to natural turnover or disease processes like cancer.

Sequencing by Synthesis

A key chemistry developed for next-generation sequencing that allows for miniaturized, highly parallel, and cyclic sequencing reactions. It enables the simultaneous reading of one base at a time across billions of DNA fragments on a small glass slide.

DNA Methylation

A chemical modification where a methyl group is added to a cytosine base, typically at CPG sites in DNA. This modification is crucial for controlling gene expression and plays a fundamental role in epigenetics, determining cell identity and state without changing the underlying DNA code.

Epigenetics

Changes in gene expression that do not involve alterations to the underlying DNA sequence, but rather modifications like DNA methylation or histone changes. It acts as the 'software of the genome,' determining which genes are expressed and when, influencing cell differentiation and function.

Sensitivity

The ability of a test to correctly identify individuals who have a particular disease (true positive rate). In cancer screening, a highly sensitive test aims to miss as few actual cancers as possible.

Specificity

The ability of a test to correctly identify individuals who do not have a particular disease (true negative rate). For broad-based cancer screening, high specificity is crucial to avoid an overwhelming number of false positives and unnecessary follow-up procedures.

Positive Predictive Value (PPV)

The probability that an individual with a positive test result actually has the disease. For cancer screening, a high PPV means that a positive test is more likely to lead to a confirmed cancer diagnosis, reducing the burden of false alarms.

Yamanaka Factors

A specific cocktail of four transcription factors discovered by Shinya Yamanaka that can reprogram fully differentiated cells, such as fibroblasts, back into a pluripotent stem cell-like state. This discovery opened new avenues for creating induced pluripotent stem cells (iPSCs) for medical research.

?
What does it mean to sequence a human genome?

Sequencing a human genome means reading out the 3 billion individual ATCG bases that comprise an individual's complete genetic code. This process has evolved from laborious, large-fragment methods to highly parallel, computational 'shotgun sequencing'.

?
How did the cost of human genome sequencing change over time?

The incremental cost of sequencing a human genome dropped from a few hundred million dollars after the initial Human Genome Project (early 2000s) to low tens of thousands of dollars by 2013, and is now around a couple hundred dollars, improving faster than Moore's Law due to miniaturization and new chemistry.

?
How was the potential for early cancer detection via liquid biopsy first discovered?

The potential was incidentally discovered during non-invasive prenatal testing (NIPT), where unusual genome-wide chromosomal abnormalities in pregnant women's cell-free DNA, not compatible with a fetus, were found to indicate the presence of undiagnosed maternal cancers.

?
Why is high specificity crucial for a universal blood test for cancer?

High specificity (low false positive rate) is critical because if a test is screening a large, relatively healthy population, even a small false positive rate (e.g., 5-10%) would lead to an unmanageable number of unnecessary and potentially harmful follow-up diagnostic procedures.

?
Which cell-free DNA feature proved most effective for multi-cancer detection and site of origin prediction?

DNA methylation patterns in cell-free DNA were found to be the most sensitive and specific feature for detecting multiple cancer types and accurately predicting their tissue of origin, outperforming mutations and chromosomal changes.

?
How does the GRAIL Galleri test predict the cancer's tissue of origin?

The test uses an atlas of methylation patterns from various healthy cell types, as methylation determines cell identity. By comparing the methylation patterns in detected cell-free tumor DNA to this atlas, the algorithm can predict which tissue or organ the cancer likely originated from.

?
Does early cancer detection through liquid biopsies translate to improved overall survival?

While definitive long-term mortality data takes decades, the biological rationale suggests that detecting cancers at earlier, localized stages (which have significantly better survival rates) and intervening early should improve outcomes. Studies are ongoing, using surrogate endpoints like reduction in stage four cancers.

?
What is the relationship between epigenetics and the biology of aging?

Epigenetic changes, particularly in DNA methylation patterns, are increasingly seen as central to aging, potentially acting as a 'hallmark of hallmarks.' These changes correlate with health status and age, suggesting that reversing them might reprogram cells back to a more youthful, higher-functioning state.

Physicians should explicitly discuss the implications of screening tests, including the positive predictive value and potential for false positives, with patients before testing. This empowers patients to understand the demands of follow-up and make informed decisions, including declining the test.

2. Prioritize Screening Specificity

When developing or evaluating population-wide screening tests, prioritize high specificity (e.g., <1% false positive rate). This prevents unnecessary follow-ups for benign conditions, which can be a significant burden.

3. Consider Multi-Cancer Blood Test

Consider adding a multi-cancer early detection blood test, such as Grail’s Gallery, to standard-of-care screening. This can double the number of cancers detected, especially for those without existing screening methods.

4. Understand Liquid Biopsy Function

Recognize that liquid biopsies are functional assays, detecting cancers actively growing, dying, and shedding DNA into the blood. This means they may not detect small, encapsulated, or indolent tumors that are not shedding DNA.

5. Negative Test, Better Prognosis

If a cancer is undetectable by a liquid biopsy, it correlates with a very good outcome, suggesting the cancer is less dangerous or existing care is effective. The test primarily detects dangerous cancers associated with higher mortality.

6. Healthy Lifestyle for Methylation

Engage in healthy lifestyle choices like not smoking, regular exercise, and maintaining a healthy weight. These behaviors are correlated with more favorable methylation patterns, which are crucial for cellular function and aging.

7. Build to Understand

Adopt the engineering principle: if you can’t build it, you don’t truly understand it. Apply this by conducting experiments to test if you can make a concept work, even in biology, to deepen your understanding.

8. First-Principles Learning

Approach problems from first principles, asking if you truly understand how something works. Design experiments that guarantee new learning about the system, regardless of hypothesis confirmation or rejection, avoiding experiments that yield no new knowledge.

9. Exhaustively Evaluate Methods

When tackling a complex problem, exhaustively evaluate every potential method or feature, rather than focusing on a single preferred approach. This ensures a comprehensive understanding and identification of the most effective solution.

10. Solve or Disprove

Approach challenging problems with the mindset that you will either figure out how to achieve the goal or definitively prove that it cannot be done. This provides clarity and direction in research and development efforts.

11. Consider NIPT for Trisomies

If pregnant, consider Non-Invasive Prenatal Testing (NIPT) to screen for trisomies (e.g., chromosome 21, 18, 13) using cell-free DNA in the blood. This highly sensitive and specific method can reduce the need for invasive procedures like amniocentesis.

12. Anticipate Rejuvenation Therapies

Stay informed about advancements in rejuvenation therapies, as experts anticipate treatments for specific organs, tissues, and cell types within the next decade. These could improve health span and resistance to diseases like cancer.

If you can't build it, you don't understand it.

Alex Aravanis

We're either going to figure out how to do this, or we're going to prove it can't be done.

Rick Klausner (attributed by Alex Aravanis)

The epigenetics [is] the software of the genome. The genetic code is kind of the hardware, but how you use it, which genes you use when, which combination, that's really the epigenetics.

Alex Aravanis

70% of people who ultimately die of cancer on their death certificate, they die from a cancer where there was no established screening prior to something like Grail's Gallery.

Alex Aravanis

The idea that we have this technology, and we're going to allow huge numbers of cancers to just progress to late stage before treating, I don't think that's the right balance of potential benefit versus burden of evidence.

Alex Aravanis
3 billion
Length of human genome Individual ATCG bases (letters) in a human's DNA code.
A few hundred million dollars
Approximate cost of an incremental human genome sequence (post-HGP) Cost immediately after the initial Human Genome Project in the early 2000s.
Low tens of thousands of dollars
Approximate cost of a whole human genome sequence (circa 2013) Cost when Alex Aravanis joined Illumina.
A couple hundred dollars
Approximate cost of a whole human genome sequence (today) Current cost for sequencing a human genome.
Around 160 base pairs
Typical length of cell-free DNA fragments This length is due to protection by nucleosomes.
About 28 million
Number of CPG sites in the human genome Positions where cytosine can be methylated.
15,000 individuals
Participants in GRAIL's CCGA study Cohort for discovering and validating cancer detection methods, approximately 60% with cancer and 40% controls.
Around 40%
GRAIL Galleri test positive predictive value (Pathfinder study) For all stages and types of cancer in a mixed-risk population (prevalence a bit less than 1%).
Less than 1%
GRAIL Galleri test false positive rate target Required for broad-based cancer screening in asymptomatic individuals.
Doubled
Detection rate of GRAIL Galleri test in Pathfinder study The number of cancers found compared to standard of care screening alone.
About 90%
Accuracy of GRAIL Galleri test for cancer site of origin For predicting the tissue where the cancer originated when a signal is detected.
About 140,000
Participants in NHS pilot study of GRAIL test Population undergoing sequential testing to look at outcomes, including reduction in stage four cancers.
70% to 80%
5-year survival for localized solid tumors When cancer has not spread to another organ.
Less than 20%
5-year survival for metastatic (stage four) solid tumors When cancer has spread to distant organs.